Daily Papers

Recently, a myriad of conditional image generation and editing models have been developed to serve different downstream tasks, including text-to-image generation, text-guided image editing, subject-driven image generation, control-guided image generation, etc. However, we observe huge inconsistencies in experimental conditions: datasets, inference, and evaluation metrics - render fair comparisons difficult. This paper proposes ImagenHub, which is a one-stop library to standardize the inference and evaluation of all the conditional image generation models. Firstly, we define seven prominent tasks and curate high-quality evaluation datasets for them. Secondly, we built a unified inference pipeline to ensure fair comparison. Thirdly, we design two human evaluation scores, i.e. Semantic Consistency and Perceptual Quality, along with comprehensive guidelines to evaluate generated images. We train expert raters to evaluate the model outputs based on the proposed metrics. Our human evaluation achieves a high inter-worker agreement of Krippendorff's alpha on 76% models with a value higher than 0.4. We comprehensively evaluated a total of around 30 models and observed three key takeaways: (1) the existing models' performance is generally unsatisfying except for Text-guided Image Generation and Subject-driven Image Generation, with 74% models achieving an overall score lower than 0.5. (2) we examined the claims from published papers and found 83% of them hold with a few exceptions. (3) None of the existing automatic metrics has a Spearman's correlation higher than 0.2 except subject-driven image generation. Moving forward, we will continue our efforts to evaluate newly published models and update our leaderboard to keep track of the progress in conditional image generation.

As Large Language Models (LLMs) saturate elementary benchmarks, the research frontier has shifted from generation to the reliability of automated evaluation. We demonstrate that standard "LLM-as-a-Judge" protocols suffer from a systematic Alignment Gap when applied to upper-undergraduate to early graduate level mathematics. To quantify this, we introduce QEDBench, the first large-scale dual-rubric alignment benchmark to systematically measure alignment with human experts on university-level math proofs by contrasting course-specific rubrics against expert common knowledge criteria. By deploying a dual-evaluation matrix (7 judges x 5 solvers) against 1,000+ hours of human evaluation, we reveal that certain frontier evaluators like Claude Opus 4.5, DeepSeek-V3, Qwen 2.5 Max, and Llama 4 Maverick exhibit significant positive bias (up to +0.18, +0.20, +0.30, +0.36 mean score inflation, respectively). Furthermore, we uncover a critical reasoning gap in the discrete domain: while Gemini 3.0 Pro achieves state-of-the-art performance (0.91 average human evaluation score), other reasoning models like GPT-5 Pro and Claude Sonnet 4.5 see their performance significantly degrade in discrete domains. Specifically, their average human evaluation scores drop to 0.72 and 0.63 in Discrete Math, and to 0.74 and 0.50 in Graph Theory. In addition to these research results, we also release QEDBench as a public benchmark for evaluating and improving AI judges. Our benchmark is publicly published at https://github.com/qqliu/Yale-QEDBench.

This paper presents the Long Context and Form Output (LCFO) benchmark, a novel evaluation framework for assessing gradual summarization and summary expansion capabilities across diverse domains. LCFO consists of long input documents (5k words average length), each of which comes with three summaries of different lengths (20%, 10%, and 5% of the input text), as well as approximately 15 questions and answers (QA) related to the input content. Notably, LCFO also provides alignments between specific QA pairs and corresponding summaries in 7 domains. The primary motivation behind providing summaries of different lengths is to establish a controllable framework for generating long texts from shorter inputs, i.e. summary expansion. To establish an evaluation metric framework for summarization and summary expansion, we provide human evaluation scores for human-generated outputs, as well as results from various state-of-the-art large language models (LLMs). GPT-4o-mini achieves best human scores among automatic systems in both summarization and summary expansion tasks (~ +10% and +20%, respectively). It even surpasses human output quality in the case of short summaries (~ +7%). Overall automatic metrics achieve low correlations with human evaluation scores (~ 0.4) but moderate correlation on specific evaluation aspects such as fluency and attribution (~ 0.6). The LCFO benchmark offers a standardized platform for evaluating summarization and summary expansion performance, as well as corresponding automatic metrics, thereby providing an important evaluation framework to advance generative AI.

Large Language Models (LLMs) have demonstrated remarkable performance on various quantitative reasoning and knowledge benchmarks. However, many of these benchmarks are losing utility as LLMs get increasingly high scores, despite not yet reaching expert performance in these domains. We introduce ARB, a novel benchmark composed of advanced reasoning problems in multiple fields. ARB presents a more challenging test than prior benchmarks, featuring problems in mathematics, physics, biology, chemistry, and law. As a subset of ARB, we introduce a challenging set of math and physics problems which require advanced symbolic reasoning and domain knowledge. We evaluate recent models such as GPT-4 and Claude on ARB and demonstrate that current models score well below 50% on more demanding tasks. In order to improve both automatic and assisted evaluation capabilities, we introduce a rubric-based evaluation approach, allowing GPT-4 to score its own intermediate reasoning steps. Further, we conduct a human evaluation of the symbolic subset of ARB, finding promising agreement between annotators and GPT-4 rubric evaluation scores.

Recent progress in transformer-based architectures has demonstrated remarkable success in video generation tasks. However, the quadratic complexity of full attention mechanisms remains a critical bottleneck, particularly for high-resolution and long-duration video sequences. In this paper, we propose NABLA, a novel Neighborhood Adaptive Block-Level Attention mechanism that dynamically adapts to sparsity patterns in video diffusion transformers (DiTs). By leveraging block-wise attention with adaptive sparsity-driven threshold, NABLA reduces computational overhead while preserving generative quality. Our method does not require custom low-level operator design and can be seamlessly integrated with PyTorch's Flex Attention operator. Experiments demonstrate that NABLA achieves up to 2.7x faster training and inference compared to baseline almost without compromising quantitative metrics (CLIP score, VBench score, human evaluation score) and visual quality drop. The code and model weights are available here: https://github.com/gen-ai-team/Wan2.1-NABLA

Existing text-to-image diffusion models struggle to synthesize realistic images given dense captions, where each text prompt provides a detailed description for a specific image region. To address this, we propose DenseDiffusion, a training-free method that adapts a pre-trained text-to-image model to handle such dense captions while offering control over the scene layout. We first analyze the relationship between generated images' layouts and the pre-trained model's intermediate attention maps. Next, we develop an attention modulation method that guides objects to appear in specific regions according to layout guidance. Without requiring additional fine-tuning or datasets, we improve image generation performance given dense captions regarding both automatic and human evaluation scores. In addition, we achieve similar-quality visual results with models specifically trained with layout conditions.

We investigate how to adapt small, efficient LLMs to Faroese, a low-resource North Germanic language. Starting from English models, we continue pre-training on related Scandinavian languages, either individually or combined via merging, before fine-tuning on Faroese. We compare full fine-tuning with parameter-efficient tuning using LoRA, evaluating their impact on both linguistic accuracy and text comprehension. Due to the lack of existing Faroese evaluation data, we construct two new minimal-pair benchmarks from adapted and newly collected datasets and complement them with human evaluations by Faroese linguists. Our results demonstrate that transfer from related languages is crucial, though the optimal source language depends on the task: Icelandic enhances linguistic accuracy, whereas Danish boosts comprehension. Similarly, the choice between full fine-tuning and LoRA is task-dependent: LoRA improves linguistic acceptability and slightly increases human evaluation scores on the base model, while full fine-tuning yields stronger comprehension performance and better preserves model capabilities during downstream fine-tuning.

Large language models show improved downstream task performance when prompted to generate step-by-step reasoning to justify their final answers. These reasoning steps greatly improve model interpretability and verification, but objectively studying their correctness (independent of the final answer) is difficult without reliable methods for automatic evaluation. We simply do not know how often the stated reasoning steps actually support the final end task predictions. In this work, we present ROSCOE, a suite of interpretable, unsupervised automatic scores that improve and extend previous text generation evaluation metrics. To evaluate ROSCOE against baseline metrics, we design a typology of reasoning errors and collect synthetic and human evaluation scores on commonly used reasoning datasets. In contrast with existing metrics, ROSCOE can measure semantic consistency, logicality, informativeness, fluency, and factuality - among other traits - by leveraging properties of step-by-step rationales. We empirically verify the strength of our metrics on five human annotated and six programmatically perturbed diagnostics datasets - covering a diverse set of tasks that require reasoning skills and show that ROSCOE can consistently outperform baseline metrics.

Recent advances in neural machine translation (NMT) have pushed the quality of machine translation systems to the point where they are becoming widely adopted to build competitive systems. However, there is still a large number of languages that are yet to reap the benefits of NMT. In this paper, we provide the first large-scale case study of the practical application of MT in the Turkic language family in order to realize the gains of NMT for Turkic languages under high-resource to extremely low-resource scenarios. In addition to presenting an extensive analysis that identifies the bottlenecks towards building competitive systems to ameliorate data scarcity, our study has several key contributions, including, i) a large parallel corpus covering 22 Turkic languages consisting of common public datasets in combination with new datasets of approximately 2 million parallel sentences, ii) bilingual baselines for 26 language pairs, iii) novel high-quality test sets in three different translation domains and iv) human evaluation scores. All models, scripts, and data will be released to the public.

Several recent end-to-end text-to-speech (TTS) models enabling single-stage training and parallel sampling have been proposed, but their sample quality does not match that of two-stage TTS systems. In this work, we present a parallel end-to-end TTS method that generates more natural sounding audio than current two-stage models. Our method adopts variational inference augmented with normalizing flows and an adversarial training process, which improves the expressive power of generative modeling. We also propose a stochastic duration predictor to synthesize speech with diverse rhythms from input text. With the uncertainty modeling over latent variables and the stochastic duration predictor, our method expresses the natural one-to-many relationship in which a text input can be spoken in multiple ways with different pitches and rhythms. A subjective human evaluation (mean opinion score, or MOS) on the LJ Speech, a single speaker dataset, shows that our method outperforms the best publicly available TTS systems and achieves a MOS comparable to ground truth.

Several recent work on speech synthesis have employed generative adversarial networks (GANs) to produce raw waveforms. Although such methods improve the sampling efficiency and memory usage, their sample quality has not yet reached that of autoregressive and flow-based generative models. In this work, we propose HiFi-GAN, which achieves both efficient and high-fidelity speech synthesis. As speech audio consists of sinusoidal signals with various periods, we demonstrate that modeling periodic patterns of an audio is crucial for enhancing sample quality. A subjective human evaluation (mean opinion score, MOS) of a single speaker dataset indicates that our proposed method demonstrates similarity to human quality while generating 22.05 kHz high-fidelity audio 167.9 times faster than real-time on a single V100 GPU. We further show the generality of HiFi-GAN to the mel-spectrogram inversion of unseen speakers and end-to-end speech synthesis. Finally, a small footprint version of HiFi-GAN generates samples 13.4 times faster than real-time on CPU with comparable quality to an autoregressive counterpart.

This paper presents a solution for the Multilingual Text Detoxification task in the PAN-2024 competition of the SmurfCat team. Using data augmentation through machine translation and a special filtering procedure, we collected an additional multilingual parallel dataset for text detoxification. Using the obtained data, we fine-tuned several multilingual sequence-to-sequence models, such as mT0 and Aya, on a text detoxification task. We applied the ORPO alignment technique to the final model. Our final model has only 3.7 billion parameters and achieves state-of-the-art results for the Ukrainian language and near state-of-the-art results for other languages. In the competition, our team achieved first place in the automated evaluation with a score of 0.52 and second place in the final human evaluation with a score of 0.74.

Obtaining meaningful quality scores for machine translation systems through human evaluation remains a challenge given the high variability between human evaluators, partly due to subjective expectations for translation quality for different language pairs. We propose a new metric called XSTS that is more focused on semantic equivalence and a cross-lingual calibration method that enables more consistent assessment. We demonstrate the effectiveness of these novel contributions in large scale evaluation studies across up to 14 language pairs, with translation both into and out of English.

Human evaluation is the gold-standard for evaluating text generation models. It is also expensive, and to fit budgetary constraints, a random subset of the test data is often chosen in practice. The randomly selected data may not accurately represent test performance, making this approach economically inefficient for model comparison. Thus, in this work, we develop a suite of selectors to get the most informative datapoints for human evaluation while taking the evaluation costs into account. We show that selectors based on variance in automated metric scores, diversity in model outputs, or Item Response Theory outperform random selection. We further develop an approach to distill these selectors to the scenario where the model outputs are not yet available. In particular, we introduce source-based estimators, which predict item usefulness for human evaluation just based on the source texts. We demonstrate the efficacy of our selectors in two common NLG tasks, machine translation and summarization, and show that up to only ~50% of the test data is needed to produce the same evaluation result as the entire data. Our implementations are published in the subset2evaluate package.

One of the goals of automatic evaluation metrics in grammatical error correction (GEC) is to rank GEC systems such that it matches human preferences. However, current automatic evaluations are based on procedures that diverge from human evaluation. Specifically, human evaluation derives rankings by aggregating sentence-level relative evaluation results, e.g., pairwise comparisons, using a rating algorithm, whereas automatic evaluation averages sentence-level absolute scores to obtain corpus-level scores, which are then sorted to determine rankings. In this study, we propose an aggregation method for existing automatic evaluation metrics which aligns with human evaluation methods to bridge this gap. We conducted experiments using various metrics, including edit-based metrics, n-gram based metrics, and sentence-level metrics, and show that resolving the gap improves results for the most of metrics on the SEEDA benchmark. We also found that even BERT-based metrics sometimes outperform the metrics of GPT-4. The proposed ranking method is integrated gec-metrics.

In this position paper, we argue that human evaluation of generative large language models (LLMs) should be a multidisciplinary undertaking that draws upon insights from disciplines such as user experience research and human behavioral psychology to ensure that the experimental design and results are reliable. The conclusions from these evaluations, thus, must consider factors such as usability, aesthetics, and cognitive biases. We highlight how cognitive biases can conflate fluent information and truthfulness, and how cognitive uncertainty affects the reliability of rating scores such as Likert. Furthermore, the evaluation should differentiate the capabilities and weaknesses of increasingly powerful large language models -- which requires effective test sets. The scalability of human evaluation is also crucial to wider adoption. Hence, to design an effective human evaluation system in the age of generative NLP, we propose the ConSiDERS-The-Human evaluation framework consisting of 6 pillars -- Consistency, Scoring Criteria, Differentiating, User Experience, Responsible, and Scalability.

Explaining the decisions of an Artificial Intelligence (AI) model is increasingly critical in many real-world, high-stake applications. Hundreds of papers have either proposed new feature attribution methods, discussed or harnessed these tools in their work. However, despite humans being the target end-users, most attribution methods were only evaluated on proxy automatic-evaluation metrics (Zhang et al. 2018; Zhou et al. 2016; Petsiuk et al. 2018). In this paper, we conduct the first user study to measure attribution map effectiveness in assisting humans in ImageNet classification and Stanford Dogs fine-grained classification, and when an image is natural or adversarial (i.e., contains adversarial perturbations). Overall, feature attribution is surprisingly not more effective than showing humans nearest training-set examples. On a harder task of fine-grained dog categorization, presenting attribution maps to humans does not help, but instead hurts the performance of human-AI teams compared to AI alone. Importantly, we found automatic attribution-map evaluation measures to correlate poorly with the actual human-AI team performance. Our findings encourage the community to rigorously test their methods on the downstream human-in-the-loop applications and to rethink the existing evaluation metrics.

We introduce SHEET, a multi-purpose open-source toolkit designed to accelerate subjective speech quality assessment (SSQA) research. SHEET stands for the Speech Human Evaluation Estimation Toolkit, which focuses on data-driven deep neural network-based models trained to predict human-labeled quality scores of speech samples. SHEET provides comprehensive training and evaluation scripts, multi-dataset and multi-model support, as well as pre-trained models accessible via Torch Hub and HuggingFace Spaces. To demonstrate its capabilities, we re-evaluated SSL-MOS, a speech self-supervised learning (SSL)-based SSQA model widely used in recent scientific papers, on an extensive list of speech SSL models. Experiments were conducted on two representative SSQA datasets named BVCC and NISQA, and we identified the optimal speech SSL model, whose performance surpassed the original SSL-MOS implementation and was comparable to state-of-the-art methods.

Large Language Models (LLMs) and other automated techniques have been increasingly used to support software developers by generating software artifacts such as code snippets, patches, and comments. However, accurately assessing the correctness of these generated artifacts remains a significant challenge. On one hand, human evaluation provides high accuracy but is labor-intensive and lacks scalability. On the other hand, many automatic evaluation metrics are scalable and require minimal human effort, but they often fail to accurately reflect the actual correctness of generated software artifacts. In this paper, we present SE-Jury, the first evaluation metric for LLM-as-Ensemble-Judge specifically designed to accurately assess the correctness of generated software artifacts. SE-Jury first defines five distinct evaluation strategies, each implemented by an independent judge. A dynamic team selection mechanism then identifies the most appropriate subset of judges as a team to produce a final correctness score through ensembling. We evaluate SE-Jury across a diverse set of software engineering (SE) benchmarks that span three popular SE tasks: code generation, automated program repair, and code summarization. Results demonstrate that SE-Jury consistently achieves a higher correlation with human judgments, with improvements ranging from 29.6% to 140.8% over existing automatic metrics. SE-Jury reaches agreement levels with human annotators that are close to inter-annotator agreement in code generation and program repair. These findings underscore SE-Jury's potential as a scalable and reliable alternative to human evaluation in these SE tasks.

Advances in diffusion, autoregressive, and hybrid models have enabled high-quality image synthesis for tasks such as text-to-image, editing, and reference-guided composition. Yet, existing benchmarks remain limited, either focus on isolated tasks, cover only narrow domains, or provide opaque scores without explaining failure modes. We introduce ImagenWorld, a benchmark of 3.6K condition sets spanning six core tasks (generation and editing, with single or multiple references) and six topical domains (artworks, photorealistic images, information graphics, textual graphics, computer graphics, and screenshots). The benchmark is supported by 20K fine-grained human annotations and an explainable evaluation schema that tags localized object-level and segment-level errors, complementing automated VLM-based metrics. Our large-scale evaluation of 14 models yields several insights: (1) models typically struggle more in editing tasks than in generation tasks, especially in local edits. (2) models excel in artistic and photorealistic settings but struggle with symbolic and text-heavy domains such as screenshots and information graphics. (3) closed-source systems lead overall, while targeted data curation (e.g., Qwen-Image) narrows the gap in text-heavy cases. (4) modern VLM-based metrics achieve Kendall accuracies up to 0.79, approximating human ranking, but fall short of fine-grained, explainable error attribution. ImagenWorld provides both a rigorous benchmark and a diagnostic tool to advance robust image generation.

The capabilities of Large Language Models (LLMs) are routinely evaluated by other LLMs trained to predict human preferences. This framework--known as LLM-as-a-judge--is highly scalable and relatively low cost. However, it is also vulnerable to malicious exploitation, as LLM responses can be tuned to overfit the preferences of the judge. Previous work shows that the answers generated by a candidate-LLM can be edited post hoc to maximise the score assigned to them by a judge-LLM. In this study, we adopt a different approach and use the signal provided by judge-LLMs as a reward to adversarially tune models that generate text preambles designed to boost downstream performance. We find that frozen LLMs pipelined with these models attain higher LLM-evaluation scores than existing frameworks. Crucially, unlike other frameworks which intervene directly on the model's response, our method is virtually undetectable. We also demonstrate that the effectiveness of the tuned preamble generator transfers when the candidate-LLM and the judge-LLM are replaced with models that are not used during training. These findings raise important questions about the design of more reliable LLM-as-a-judge evaluation settings. They also demonstrate that human preferences can be reverse engineered effectively, by pipelining LLMs to optimise upstream preambles via reinforcement learning--an approach that could find future applications in diverse tasks and domains beyond adversarial attacks.

Evaluating the factuality of long-form text generated by large language models (LMs) is non-trivial because (1) generations often contain a mixture of supported and unsupported pieces of information, making binary judgments of quality inadequate, and (2) human evaluation is time-consuming and costly. In this paper, we introduce FActScore (Factual precision in Atomicity Score), a new evaluation that breaks a generation into a series of atomic facts and computes the percentage of atomic facts supported by a reliable knowledge source. We conduct an extensive human evaluation to obtain FActScores of people biographies generated by several state-of-the-art commercial LMs -- InstructGPT, ChatGPT, and the retrieval-augmented PerplexityAI -- and report new analysis demonstrating the need for such a fine-grained score (e.g., ChatGPT only achieves 58%). Since human evaluation is costly, we also introduce an automated model that estimates FActScore, using retrieval and a strong language model, with less than a 2% error rate. Finally, we use this automated metric to evaluate 6,500 generations from a new set of 13 recent LMs that would have cost $26K if evaluated by humans, with various findings: GPT-4 and ChatGPT are more factual than public models, and Vicuna and Alpaca are some of the best public models.

This paper addresses the problem of evaluating the quality of automatically generated subtitles, which includes not only the quality of the machine-transcribed or translated speech, but also the quality of line segmentation and subtitle timing. We propose SubER - a single novel metric based on edit distance with shifts that takes all of these subtitle properties into account. We compare it to existing metrics for evaluating transcription, translation, and subtitle quality. A careful human evaluation in a post-editing scenario shows that the new metric has a high correlation with the post-editing effort and direct human assessment scores, outperforming baseline metrics considering only the subtitle text, such as WER and BLEU, and existing methods to integrate segmentation and timing features.

Neural machine translation (NMT) has progressed rapidly in the past few years, promising improvements and quality translations for different languages. Evaluation of this task is crucial to determine the quality of the translation. Overall, insufficient emphasis is placed on the actual sense of the translation in traditional methods. We propose a bidirectional semantic-based evaluation method designed to assess the sense distance of the translation from the source text. This approach employs the comprehensive multilingual encyclopedic dictionary BabelNet. Through the calculation of the semantic distance between the source and its back translation of the output, our method introduces a quantifiable approach that empowers sentence comparison on the same linguistic level. Factual analysis shows a strong correlation between the average evaluation scores generated by our method and the human assessments across various machine translation systems for English-German language pair. Finally, our method proposes a new multilingual approach to rank MT systems without the need for parallel corpora.

Commit messages are natural language descriptions of code changes, which are important for program understanding and maintenance. However, writing commit messages manually is time-consuming and laborious, especially when the code is updated frequently. Various approaches utilizing generation or retrieval techniques have been proposed to automatically generate commit messages. To achieve a better understanding of how the existing approaches perform in solving this problem, this paper conducts a systematic and in-depth analysis of the state-of-the-art models and datasets. We find that: (1) Different variants of the BLEU metric are used in previous works, which affects the evaluation and understanding of existing methods. (2) Most existing datasets are crawled only from Java repositories while repositories in other programming languages are not sufficiently explored. (3) Dataset splitting strategies can influence the performance of existing models by a large margin. Some models show better performance when the datasets are split by commit, while other models perform better when the datasets are split by timestamp or by project. Based on our findings, we conduct a human evaluation and find the BLEU metric that best correlates with the human scores for the task. We also collect a large-scale, information-rich, and multi-language commit message dataset MCMD and evaluate existing models on this dataset. Furthermore, we conduct extensive experiments under different dataset splitting strategies and suggest the suitable models under different scenarios. Based on the experimental results and findings, we provide feasible suggestions for comprehensively evaluating commit message generation models and discuss possible future research directions. We believe this work can help practitioners and researchers better evaluate and select models for automatic commit message generation.

The rapid advancement of AI-generated video models has created a pressing need for robust and interpretable evaluation frameworks. Existing metrics are limited to producing numerical scores without explanatory comments, resulting in low interpretability and human evaluation alignment. To address those challenges, we introduce AIGVE-MACS, a unified model for AI-Generated Video Evaluation(AIGVE), which can provide not only numerical scores but also multi-aspect language comment feedback in evaluating these generated videos. Central to our approach is AIGVE-BENCH 2, a large-scale benchmark comprising 2,500 AI-generated videos and 22,500 human-annotated detailed comments and numerical scores across nine critical evaluation aspects. Leveraging AIGVE-BENCH 2, AIGVE-MACS incorporates recent Vision-Language Models with a novel token-wise weighted loss and a dynamic frame sampling strategy to better align with human evaluators. Comprehensive experiments across supervised and zero-shot benchmarks demonstrate that AIGVE-MACS achieves state-of-the-art performance in both scoring correlation and comment quality, significantly outperforming prior baselines including GPT-4o and VideoScore. In addition, we further showcase a multi-agent refinement framework where feedback from AIGVE-MACS drives iterative improvements in video generation, leading to 53.5% quality enhancement. This work establishes a new paradigm for comprehensive, human-aligned evaluation of AI-generated videos. We release the AIGVE-BENCH 2 and AIGVE-MACS at https://huggingface.co/xiaoliux/AIGVE-MACS.

We present Meena, a multi-turn open-domain chatbot trained end-to-end on data mined and filtered from public domain social media conversations. This 2.6B parameter neural network is simply trained to minimize perplexity of the next token. We also propose a human evaluation metric called Sensibleness and Specificity Average (SSA), which captures key elements of a human-like multi-turn conversation. Our experiments show strong correlation between perplexity and SSA. The fact that the best perplexity end-to-end trained Meena scores high on SSA (72% on multi-turn evaluation) suggests that a human-level SSA of 86% is potentially within reach if we can better optimize perplexity. Additionally, the full version of Meena (with a filtering mechanism and tuned decoding) scores 79% SSA, 23% higher in absolute SSA than the existing chatbots we evaluated.

Virtual film production requires intricate decision-making processes, including scriptwriting, virtual cinematography, and precise actor positioning and actions. Motivated by recent advances in automated decision-making with language agent-based societies, this paper introduces FilmAgent, a novel LLM-based multi-agent collaborative framework for end-to-end film automation in our constructed 3D virtual spaces. FilmAgent simulates various crew roles, including directors, screenwriters, actors, and cinematographers, and covers key stages of a film production workflow: (1) idea development transforms brainstormed ideas into structured story outlines; (2) scriptwriting elaborates on dialogue and character actions for each scene; (3) cinematography determines the camera setups for each shot. A team of agents collaborates through iterative feedback and revisions, thereby verifying intermediate scripts and reducing hallucinations. We evaluate the generated videos on 15 ideas and 4 key aspects. Human evaluation shows that FilmAgent outperforms all baselines across all aspects and scores 3.98 out of 5 on average, showing the feasibility of multi-agent collaboration in filmmaking. Further analysis reveals that FilmAgent, despite using the less advanced GPT-4o model, surpasses the single-agent o1, showing the advantage of a well-coordinated multi-agent system. Lastly, we discuss the complementary strengths and weaknesses of OpenAI's text-to-video model Sora and our FilmAgent in filmmaking.

Conversational Information Retrieval (CIR) systems, while offering intuitive access to information, face a significant challenge: reliably handling unanswerable questions to prevent the generation of misleading or hallucinated content. Traditional approaches often rely on external classifiers, which can introduce inconsistencies with the core generative Large Language Models (LLMs). This paper introduces Self-Aware LLM for Unanswerability (SALU), a novel approach that deeply integrates unanswerability detection directly within the LLM's generative process. SALU is trained using a multi-task learning framework for both standard Question Answering (QA) and explicit abstention generation for unanswerable queries. Crucially, it incorporates a confidence-score-guided reinforcement learning with human feedback (RLHF) phase, which explicitly penalizes hallucinated responses and rewards appropriate abstentions, fostering intrinsic self-awareness of knowledge boundaries. Through extensive experiments on our custom-built C-IR_Answerability dataset, SALU consistently outperforms strong baselines, including hybrid LLM-classifier systems, in overall accuracy for correctly answering or abstaining from questions. Human evaluation further confirms SALU's superior reliability, achieving high scores in factuality, appropriate abstention, and, most importantly, a dramatic reduction in hallucination, demonstrating its ability to robustly "know when to say 'I don't know'."

Video generation models have achieved remarkable progress in the past year. The quality of AI video continues to improve, but at the cost of larger model size, increased data quantity, and greater demand for training compute. In this report, we present Open-Sora 2.0, a commercial-level video generation model trained for only $200k. With this model, we demonstrate that the cost of training a top-performing video generation model is highly controllable. We detail all techniques that contribute to this efficiency breakthrough, including data curation, model architecture, training strategy, and system optimization. According to human evaluation results and VBench scores, Open-Sora 2.0 is comparable to global leading video generation models including the open-source HunyuanVideo and the closed-source Runway Gen-3 Alpha. By making Open-Sora 2.0 fully open-source, we aim to democratize access to advanced video generation technology, fostering broader innovation and creativity in content creation. All resources are publicly available at: https://github.com/hpcaitech/Open-Sora.

Biomedical literature often uses complex language and inaccessible professional terminologies. That is why simplification plays an important role in improving public health literacy. Applying Natural Language Processing (NLP) models to automate such tasks allows for quick and direct accessibility for lay readers. In this work, we investigate the ability of state-of-the-art large language models (LLMs) on the task of biomedical abstract simplification, using the publicly available dataset for plain language adaptation of biomedical abstracts (PLABA). The methods applied include domain fine-tuning and prompt-based learning (PBL) on: 1) Encoder-decoder models (T5, SciFive, and BART), 2) Decoder-only GPT models (GPT-3.5 and GPT-4) from OpenAI and BioGPT, and 3) Control-token mechanisms on BART-based models. We used a range of automatic evaluation metrics, including BLEU, ROUGE, SARI, and BERTscore, and also conducted human evaluations. BART-Large with Control Token (BART-L-w-CT) mechanisms reported the highest SARI score of 46.54 and T5-base reported the highest BERTscore 72.62. In human evaluation, BART-L-w-CTs achieved a better simplicity score over T5-Base (2.9 vs. 2.2), while T5-Base achieved a better meaning preservation score over BART-L-w-CTs (3.1 vs. 2.6). We also categorised the system outputs with examples, hoping this will shed some light for future research on this task. Our code, fine-tuned models, and data splits are available at https://github.com/HECTA-UoM/PLABA-MU

Recently, denoising diffusion probabilistic models and generative score matching have shown high potential in modelling complex data distributions while stochastic calculus has provided a unified point of view on these techniques allowing for flexible inference schemes. In this paper we introduce Grad-TTS, a novel text-to-speech model with score-based decoder producing mel-spectrograms by gradually transforming noise predicted by encoder and aligned with text input by means of Monotonic Alignment Search. The framework of stochastic differential equations helps us to generalize conventional diffusion probabilistic models to the case of reconstructing data from noise with different parameters and allows to make this reconstruction flexible by explicitly controlling trade-off between sound quality and inference speed. Subjective human evaluation shows that Grad-TTS is competitive with state-of-the-art text-to-speech approaches in terms of Mean Opinion Score. We will make the code publicly available shortly.

Contemporary image generation systems have achieved high fidelity and superior aesthetic quality beyond basic text-image alignment. However, existing evaluation frameworks have failed to evolve in parallel. This study reveals that human preference reward models fine-tuned based on CLIP and BLIP architectures have inherent flaws: they inappropriately assign low scores to images with rich details and high aesthetic value, creating a significant discrepancy with actual human aesthetic preferences. To address this issue, we design a novel evaluation score, ICT (Image-Contained-Text) score, that achieves and surpasses the objectives of text-image alignment by assessing the degree to which images represent textual content. Building upon this foundation, we further train an HP (High-Preference) score model using solely the image modality to enhance image aesthetics and detail quality while maintaining text-image alignment. Experiments demonstrate that the proposed evaluation model improves scoring accuracy by over 10\% compared to existing methods, and achieves significant results in optimizing state-of-the-art text-to-image models. This research provides theoretical and empirical support for evolving image generation technology toward higher-order human aesthetic preferences. Code is available at https://github.com/BarretBa/ICTHP.

Retrieval-augmented generation (RAG) has emerged as a promising solution to mitigate the limitations of large language models (LLMs), such as hallucinations and outdated information. However, it remains unclear how LLMs handle knowledge conflicts arising from different augmented retrieved passages, especially when these passages originate from the same source and have equal trustworthiness. In this work, we conduct a comprehensive evaluation of LLM-generated answers to questions that have varying answers based on contradictory passages from Wikipedia, a dataset widely regarded as a high-quality pre-training resource for most LLMs. Specifically, we introduce WikiContradict, a benchmark consisting of 253 high-quality, human-annotated instances designed to assess LLM performance when augmented with retrieved passages containing real-world knowledge conflicts. We benchmark a diverse range of both closed and open-source LLMs under different QA scenarios, including RAG with a single passage, and RAG with 2 contradictory passages. Through rigorous human evaluations on a subset of WikiContradict instances involving 5 LLMs and over 3,500 judgements, we shed light on the behaviour and limitations of these models. For instance, when provided with two passages containing contradictory facts, all models struggle to generate answers that accurately reflect the conflicting nature of the context, especially for implicit conflicts requiring reasoning. Since human evaluation is costly, we also introduce an automated model that estimates LLM performance using a strong open-source language model, achieving an F-score of 0.8. Using this automated metric, we evaluate more than 1,500 answers from seven LLMs across all WikiContradict instances. To facilitate future work, we release WikiContradict on: https://ibm.biz/wikicontradict.

Building open-domain conversational systems (or chatbots) that produce convincing responses is a recognized challenge. Recent state-of-the-art (SoTA) transformer-based models for the generation of natural language dialogue have demonstrated impressive performance in simulating human-like, single-turn conversations in English. This work investigates, by an empirical study, the potential for transfer learning of such models to Swedish language. DialoGPT, an English language pre-trained model, is adapted by training on three different Swedish language conversational datasets obtained from publicly available sources. Perplexity score (an automated intrinsic language model metric) and surveys by human evaluation were used to assess the performances of the fine-tuned models, with results that indicate that the capacity for transfer learning can be exploited with considerable success. Human evaluators asked to score the simulated dialogue judged over 57% of the chatbot responses to be human-like for the model trained on the largest (Swedish) dataset. We provide the demos and model checkpoints of our English and Swedish chatbots on the HuggingFace platform for public use.

Large Language Models (LLMs) have demonstrated impressive performance across diverse domains, yet they still encounter challenges such as insufficient domain-specific knowledge, biases, and hallucinations. This underscores the need for robust evaluation methodologies to accurately assess LLM-based applications. Traditional evaluation methods, which rely on word overlap or text embeddings, are inadequate for capturing the nuanced semantic information necessary to evaluate dynamic, open-ended text generation. Recent research has explored leveraging LLMs to mimic human reasoning and decision-making processes for evaluation purposes known as LLM-as-a-judge framework. However, these existing frameworks have two significant limitations. First, they lack the flexibility to adapt to different text styles, including various answer and ground truth styles, thereby reducing their generalization performance. Second, the evaluation scores produced by these frameworks are often skewed and hard to interpret, showing a low correlation with human judgment. To address these challenges, we propose a novel dynamic multi-agent system that automatically designs personalized LLM judges for various natural language generation applications. This system iteratively refines evaluation prompts and balances the trade-off between the adaptive requirements of downstream tasks and the alignment with human perception. Our experimental results show that the proposed multi-agent LLM Judge framework not only enhances evaluation accuracy compared to existing methods but also produces evaluation scores that better align with human perception.

GPT-3 and GPT-4 models are powerful, achieving high performance on a variety of Natural Language Processing tasks. However, there is a relative lack of detailed published analysis of their performance on the task of grammatical error correction (GEC). To address this, we perform experiments testing the capabilities of a GPT-3.5 model (text-davinci-003) and a GPT-4 model (gpt-4-0314) on major GEC benchmarks. We compare the performance of different prompts in both zero-shot and few-shot settings, analyzing intriguing or problematic outputs encountered with different prompt formats. We report the performance of our best prompt on the BEA-2019 and JFLEG datasets, finding that the GPT models can perform well in a sentence-level revision setting, with GPT-4 achieving a new high score on the JFLEG benchmark. Through human evaluation experiments, we compare the GPT models' corrections to source, human reference, and baseline GEC system sentences and observe differences in editing strategies and how they are scored by human raters.

Persona-driven role-playing (PRP) aims to build AI characters that can respond to user queries by faithfully sticking with all persona statements. Unfortunately, existing faithfulness criteria for PRP are limited to coarse-grained LLM-based scoring without a clear definition or formulation. This paper presents a pioneering exploration to quantify PRP faithfulness as a fine-grained and explainable criterion, which also serves as a reliable reference for optimization. Our criterion first discriminates persona statements into active and passive constraints by identifying the query-statement relevance. Then, we incorporate all constraints following the principle that the AI character's response should be (a) entailed by active (relevant) constraints and (b) not contradicted by passive (irrelevant) constraints. We translate this principle mathematically into a novel Active-Passive-Constraint (APC) score, a constraint-wise sum of natural language inference (NLI) scores weighted by relevance scores. In practice, we build the APC scoring system by symbolically distilling small discriminators from GPT-4 for efficiency. We validate the quality of the APC score against human evaluation based on example personas with tens of statements, and the results show a high correlation. We further leverage it as a reward system in direct preference optimization (DPO) for better AI characters. Our experiments offer a fine-grained and explainable comparison between existing PRP techniques, revealing their advantages and limitations. We further find APC-based DPO to be one of the most competitive techniques for sticking with all constraints and can be well incorporated with other techniques. We then extend the scale of the experiments to real persons with hundreds of statements and reach a consistent conclusion.

State-of-the-art speech synthesis models try to get as close as possible to the human voice. Hence, modelling emotions is an essential part of Text-To-Speech (TTS) research. In our work, we selected FastSpeech2 as the starting point and proposed a series of modifications for synthesizing emotional speech. According to automatic and human evaluation, our model, EmoSpeech, surpasses existing models regarding both MOS score and emotion recognition accuracy in generated speech. We provided a detailed ablation study for every extension to FastSpeech2 architecture that forms EmoSpeech. The uneven distribution of emotions in the text is crucial for better, synthesized speech and intonation perception. Our model includes a conditioning mechanism that effectively handles this issue by allowing emotions to contribute to each phone with varying intensity levels. The human assessment indicates that proposed modifications generate audio with higher MOS and emotional expressiveness.

Charts are commonly used for exploring data and communicating insights. Generating natural language summaries from charts can be very helpful for people in inferring key insights that would otherwise require a lot of cognitive and perceptual efforts. We present Chart-to-text, a large-scale benchmark with two datasets and a total of 44,096 charts covering a wide range of topics and chart types. We explain the dataset construction process and analyze the datasets. We also introduce a number of state-of-the-art neural models as baselines that utilize image captioning and data-to-text generation techniques to tackle two problem variations: one assumes the underlying data table of the chart is available while the other needs to extract data from chart images. Our analysis with automatic and human evaluation shows that while our best models usually generate fluent summaries and yield reasonable BLEU scores, they also suffer from hallucinations and factual errors as well as difficulties in correctly explaining complex patterns and trends in charts.

Professionals working in technical domain typically hand-draw (on whiteboard, paper, etc.) technical diagrams (e.g., flowcharts, block diagrams, etc.) during discussions; however, if they want to edit these later, it needs to be drawn from scratch. Modern day VLMs have made tremendous progress in image understanding but they struggle when it comes to understanding technical diagrams. One way to overcome this problem is to fine-tune on real world hand-drawn images, but it is not practically possible to generate large number of such images. In this paper, we introduce a large synthetically generated corpus (reflective of real world images) for training VLMs and subsequently evaluate VLMs on a smaller corpus of hand-drawn images (with the help of humans). We introduce several new self-supervision tasks for training and perform extensive experiments with various baseline models and fine-tune Llama 3.2 11B-instruct model on synthetic images on these tasks to obtain LLama-VL-TUG, which significantly improves the ROUGE-L performance of Llama 3.2 11B-instruct by 2.14x and achieves the best all-round performance across all baseline models. On real-world images, human evaluation reveals that we achieve minimum compilation errors across all baselines in 7 out of 8 diagram types and improve the average F1 score of Llama 3.2 11B-instruct by 6.97x.

Effective communication, specifically through documentation, is the beating heart of collaboration among contributors in software development. Recent advancements in language models (LMs) have enabled the introduction of a new type of actor in that ecosystem: LM-powered assistants capable of code generation, optimization, and maintenance. Our study investigates the efficacy of small language models (SLMs) for generating high-quality docstrings by assessing accuracy, conciseness, and clarity, benchmarking performance quantitatively through mathematical formulas and qualitatively through human evaluation using Likert scale. Further, we introduce DocuMint, as a large-scale supervised fine-tuning dataset with 100,000 samples. In quantitative experiments, Llama 3 8B achieved the best performance across all metrics, with conciseness and clarity scores of 0.605 and 64.88, respectively. However, under human evaluation, CodeGemma 7B achieved the highest overall score with an average of 8.3 out of 10 across all metrics. Fine-tuning the CodeGemma 2B model using the DocuMint dataset led to significant improvements in performance across all metrics, with gains of up to 22.5% in conciseness. The fine-tuned model and the dataset can be found in HuggingFace and the code can be found in the repository.

Query Auto-Completion (QAC) suggests query completions as users type, helping them articulate intent and reach results more efficiently. Existing approaches face fundamental challenges: traditional retrieve-and-rank pipelines have limited long-tail coverage and require extensive feature engineering, while recent generative methods suffer from hallucination and safety risks. We present a unified framework that reformulates QAC as end-to-end list generation through Retrieval-Augmented Generation (RAG) and multi-objective Direct Preference Optimization (DPO). Our approach combines three key innovations: (1) reformulating QAC as end-to-end list generation with multi-objective optimization; (2) defining and deploying a suite of rule-based, model-based, and LLM-as-judge verifiers for QAC, and using them in a comprehensive methodology that combines RAG, multi-objective DPO, and iterative critique-revision for high-quality synthetic data; (3) a hybrid serving architecture enabling efficient production deployment under strict latency constraints. Evaluation on a large-scale commercial search platform demonstrates substantial improvements: offline metrics show gains across all dimensions, human evaluation yields +0.40 to +0.69 preference scores, and a controlled online experiment achieves 5.44\% reduction in keystrokes and 3.46\% increase in suggestion adoption, validating that unified generation with RAG and multi-objective alignment provides an effective solution for production QAC. This work represents a paradigm shift to end-to-end generation powered by large language models, RAG, and multi-objective alignment, establishing a production-validated framework that can benefit the broader search and recommendation industry.

LLM-as-a-judge is a framework in which a large language model (LLM) automatically evaluates the output of another LLM. We propose quantitative LLM judges, which align evaluation scores of existing LLM judges to human scores in a given domain using regression models. The models are trained to improve the score of the original judge by using the judge's textual evaluation and score. We present four quantitative judges for different types of absolute and relative feedback, which showcases the generality and versatility of our framework. Our framework is more computationally efficient than supervised fine-tuning and can be more statistically efficient when human feedback is limited, which is expected in most applications of our work. We validate these claims empirically on four datasets using two base judges. Our experiments show that quantitative judges can effectively improve the predictive power of existing judges through post-hoc modeling.

We report the results of DialogSum Challenge, the shared task on summarizing real-life scenario dialogues at INLG 2022. Four teams participate in this shared task and three submit their system reports, exploring different methods to improve the performance of dialogue summarization. Although there is a great improvement over the baseline models regarding automatic evaluation metrics, such as Rouge scores, we find that there is a salient gap between model generated outputs and human annotated summaries by human evaluation from multiple aspects. These findings demonstrate the difficulty of dialogue summarization and suggest that more fine-grained evaluatuion metrics are in need.

Generative adversarial networks have seen rapid development in recent years and have led to remarkable improvements in generative modelling of images. However, their application in the audio domain has received limited attention, and autoregressive models, such as WaveNet, remain the state of the art in generative modelling of audio signals such as human speech. To address this paucity, we introduce GAN-TTS, a Generative Adversarial Network for Text-to-Speech. Our architecture is composed of a conditional feed-forward generator producing raw speech audio, and an ensemble of discriminators which operate on random windows of different sizes. The discriminators analyse the audio both in terms of general realism, as well as how well the audio corresponds to the utterance that should be pronounced. To measure the performance of GAN-TTS, we employ both subjective human evaluation (MOS - Mean Opinion Score), as well as novel quantitative metrics (Fr\'echet DeepSpeech Distance and Kernel DeepSpeech Distance), which we find to be well correlated with MOS. We show that GAN-TTS is capable of generating high-fidelity speech with naturalness comparable to the state-of-the-art models, and unlike autoregressive models, it is highly parallelisable thanks to an efficient feed-forward generator. Listen to GAN-TTS reading this abstract at https://storage.googleapis.com/deepmind-media/research/abstract.wav.

Current medical question answering systems have difficulty processing long, detailed and informally worded questions submitted by patients, called Consumer Health Questions (CHQs). To address this issue, we introduce a medical question understanding and answering system with knowledge grounding and semantic self-supervision. Our system is a pipeline that first summarizes a long, medical, user-written question, using a supervised summarization loss. Then, our system performs a two-step retrieval to return answers. The system first matches the summarized user question with an FAQ from a trusted medical knowledge base, and then retrieves a fixed number of relevant sentences from the corresponding answer document. In the absence of labels for question matching or answer relevance, we design 3 novel, self-supervised and semantically-guided losses. We evaluate our model against two strong retrieval-based question answering baselines. Evaluators ask their own questions and rate the answers retrieved by our baselines and own system according to their relevance. They find that our system retrieves more relevant answers, while achieving speeds 20 times faster. Our self-supervised losses also help the summarizer achieve higher scores in ROUGE, as well as in human evaluation metrics. We release our code to encourage further research.

This paper investigates whether large language models (LLMs) are state-of-the-art quality estimators for machine translation of user-generated content (UGC) that contains emotional expressions, without the use of reference translations. To achieve this, we employ an existing emotion-related dataset with human-annotated errors and calculate quality evaluation scores based on the Multi-dimensional Quality Metrics. We compare the accuracy of several LLMs with that of our fine-tuned baseline models, under in-context learning and parameter-efficient fine-tuning (PEFT) scenarios. We find that PEFT of LLMs leads to better performance in score prediction with human interpretable explanations than fine-tuned models. However, a manual analysis of LLM outputs reveals that they still have problems such as refusal to reply to a prompt and unstable output while evaluating machine translation of UGC.

Recent advancements in machine learning have spurred growing interests in automated interpreting quality assessment. Nevertheless, existing research suffers from insufficient examination of language use quality, unsatisfactory modeling effectiveness due to data scarcity and imbalance, and a lack of efforts to explain model predictions. To address these gaps, we propose a multi-dimensional modeling framework that integrates feature engineering, data augmentation, and explainable machine learning. This approach prioritizes explainability over ``black box'' predictions by utilizing only construct-relevant, transparent features and conducting Shapley Value (SHAP) analysis. Our results demonstrate strong predictive performance on a novel English-Chinese consecutive interpreting dataset, identifying BLEURT and CometKiwi scores to be the strongest predictive features for fidelity, pause-related features for fluency, and Chinese-specific phraseological diversity metrics for language use. Overall, by placing particular emphasis on explainability, we present a scalable, reliable, and transparent alternative to traditional human evaluation, facilitating the provision of detailed diagnostic feedback for learners and supporting self-regulated learning advantages not afforded by automated scores in isolation.

Recent great advances in video generation models have demonstrated their potential to produce high-quality videos, bringing challenges to effective evaluation. Unlike human evaluation, existing automated evaluation metrics lack high-level semantic understanding and reasoning capabilities for video, thus making them infeasible and unexplainable. To fill this gap, we curate GRADEO-Instruct, a multi-dimensional T2V evaluation instruction tuning dataset, including 3.3k videos from over 10 existing video generation models and multi-step reasoning assessments converted by 16k human annotations. We then introduce GRADEO, one of the first specifically designed video evaluation models, which grades AI-generated videos for explainable scores and assessments through multi-step reasoning. Experiments show that our method aligns better with human evaluations than existing methods. Furthermore, our benchmarking reveals that current video generation models struggle to produce content that aligns with human reasoning and complex real-world scenarios. The models, datasets, and codes will be released soon.

We introduce a Depicted image Quality Assessment method (DepictQA), overcoming the constraints of traditional score-based approaches. DepictQA leverages Multi-modal Large Language Models (MLLMs), allowing for detailed, language-based, human-like evaluation of image quality. Unlike conventional Image Quality Assessment (IQA) methods relying on scores, DepictQA interprets image content and distortions descriptively and comparatively, aligning closely with humans' reasoning process. To build the DepictQA model, we establish a hierarchical task framework, and collect a multi-modal IQA training dataset, named M-BAPPS. To navigate the challenges in limited training data and processing multiple images, we propose to use multi-source training data and specialized image tags. Our DepictQA demonstrates a better performance than score-based methods on the BAPPS benchmark. Moreover, compared with general MLLMs, our DepictQA can generate more accurate reasoning descriptive languages. Our research indicates that language-based IQA methods have the potential to be customized for individual preferences. Datasets and codes will be released publicly.

We introduce KoBALT (Korean Benchmark for Advanced Linguistic Tasks), a comprehensive linguistically-motivated benchmark comprising 700 multiple-choice questions spanning 24 phenomena across five linguistic domains: syntax, semantics, pragmatics, phonetics/phonology, and morphology. KoBALT is designed to advance the evaluation of large language models (LLMs) in Korean, a morphologically rich language, by addressing the limitations of conventional benchmarks that often lack linguistic depth and typological grounding. It introduces a suite of expert-curated, linguistically motivated questions with minimal n-gram overlap with standard Korean corpora, substantially mitigating the risk of data contamination and allowing a more robust assessment of true language understanding. Our evaluation of 20 contemporary LLMs reveals significant performance disparities, with the highest-performing model achieving 61\% general accuracy but showing substantial variation across linguistic domains - from stronger performance in semantics (66\%) to considerable weaknesses in phonology (31\%) and morphology (36\%). Through human preference evaluation with 95 annotators, we demonstrate a strong correlation between KoBALT scores and human judgments, validating our benchmark's effectiveness as a discriminative measure of Korean language understanding. KoBALT addresses critical gaps in linguistic evaluation for typologically diverse languages and provides a robust framework for assessing genuine linguistic competence in Korean language models.

Evaluating generative foundation models on open-ended multimodal understanding (MMU) and generation (MMG) tasks across diverse modalities (e.g., images, audio, video) poses significant challenges due to the complexity of cross-modal interactions. To this end, the idea of utilizing Multimodal LLMs (MLLMs) as automated judges has emerged, with encouraging results in assessing vision-language understanding tasks. Moving further, this paper extends MLLM-as-a-Judge across modalities to a unified manner by introducing two benchmarks, TaskAnything and JudgeAnything, to respectively evaluate the overall performance and judging capabilities of MLLMs across any-to-any modality tasks. Specifically, TaskAnything evaluates the MMU and MMG capabilities across 15 any-to-any modality categories, employing 1,500 queries curated from well-established benchmarks. Furthermore, JudgeAnything evaluates the judging capabilities of 5 advanced (e.g., GPT-4o and Gemini-2.0-Flash) from the perspectives of Pair Comparison and Score Evaluation, providing a standardized testbed that incorporates human judgments and detailed rubrics. Our extensive experiments reveal that while these MLLMs show promise in assessing MMU (i.e., achieving an average of 66.55% in Pair Comparison setting and 42.79% in Score Evaluation setting), they encounter significant challenges with MMG tasks (i.e., averaging only 53.37% in Pair Comparison setting and 30.05% in Score Evaluation setting), exposing cross-modality biases and hallucination issues. To address this, we present OmniArena, an automated platform for evaluating omni-models and multimodal reward models. Our work highlights the need for fairer evaluation protocols and stronger alignment with human preferences. The source code and dataset are publicly available at: https://urrealhero.github.io/judgeanythingweb/.

Charts are central to analytical reasoning, yet existing benchmarks for chart understanding focus almost exclusively on single-chart interpretation rather than comparative reasoning across multiple charts. To address this gap, we introduce ChartDiff, the first large-scale benchmark for cross-chart comparative summarization. ChartDiff consists of 8,541 chart pairs spanning diverse data sources, chart types, and visual styles, each annotated with LLM-generated and human-verified summaries describing differences in trends, fluctuations, and anomalies. Using ChartDiff, we evaluate general-purpose, chart-specialized, and pipeline-based models. Our results show that frontier general-purpose models achieve the highest GPT-based quality, while specialized and pipeline-based methods obtain higher ROUGE scores but lower human-aligned evaluation, revealing a clear mismatch between lexical overlap and actual summary quality. We further find that multi-series charts remain challenging across model families, whereas strong end-to-end models are relatively robust to differences in plotting libraries. Overall, our findings demonstrate that comparative chart reasoning remains a significant challenge for current vision-language models and position ChartDiff as a new benchmark for advancing research on multi-chart understanding.

Large Language Models (LLMs) have catalyzed vibe coding, where users leverage LLMs to generate and iteratively refine code through natural language interactions until it passes their vibe check. Vibe check is tied to real-world human preference and goes beyond functionality: the solution should feel right, read cleanly, preserve intent, and remain correct. However, current code evaluation remains anchored to pass@k and captures only functional correctness, overlooking the non-functional instructions that users routinely apply. In this paper, we hypothesize that instruction following is the missing piece underlying vibe check that represents human preference in coding besides functional correctness. To quantify models' code instruction following capabilities with measurable signals, we present VeriCode, a taxonomy of 30 verifiable code instructions together with corresponding deterministic verifiers. We use the taxonomy to augment established evaluation suites, resulting in Vibe Checker, a testbed to assess both code instruction following and functional correctness. Upon evaluating 31 leading LLMs, we show that even the strongest models struggle to comply with multiple instructions and exhibit clear functional regression. Most importantly, a composite score of functional correctness and instruction following correlates the best with human preference, with the latter emerging as the primary differentiator on real-world programming tasks. Our work identifies core factors of the vibe check, providing a concrete path for benchmarking and developing models that better align with user preferences in coding.

While prior work has examined the generation capabilities of Agentic AI systems, little is known about how reviewers respond to AI-authored code in practice. In this paper, we present a large-scale empirical study of code review dynamics in agent-generated PRs. Using a curated subset of the AIDev dataset, we analyze 19,450 inline review comments spanning 3,177 agent-authored PRs from real-world GitHub repositories. We first derive a taxonomy of 12 review comment themes using topic modeling combined with large language model (LLM)-assisted semantic clustering and consolidation. According to this taxonomy, we then investigate whether zero-shot prompts to LLM can reliably annotate review comments. Our evaluation against human annotations shows that open-source LLM achieves reasonably high exact match (78.63%), macro F1 score (0.78), and substantial agreement with human annotators at the review comment level. At the PR level, the LLM also correctly identifies the dominant review theme with 78% Top-1 accuracy and achieves an average Jaccard similarity of 0.76, indicating strong alignment with human judgments. Applying this annotation pipeline at scale, we find that apart from functional correctness and logical changes, reviews of agent-authored PRs predominantly focus on documentation gaps, refactoring needs, styling and formatting issues, with testing and security-related concerns. These findings suggest that while AI agents can accelerate code production, there remain gaps requiring targeted human review oversight.

Recent text-to-image generative models can generate high-fidelity images from text inputs, but the quality of these generated images cannot be accurately evaluated by existing evaluation metrics. To address this issue, we introduce Human Preference Dataset v2 (HPD v2), a large-scale dataset that captures human preferences on images from a wide range of sources. HPD v2 comprises 798,090 human preference choices on 430,060 pairs of images, making it the largest dataset of its kind. The text prompts and images are deliberately collected to eliminate potential bias, which is a common issue in previous datasets. By fine-tuning CLIP on HPD v2, we obtain Human Preference Score v2 (HPS v2), a scoring model that can more accurately predict text-generated images' human preferences. Our experiments demonstrate that HPS v2 generalizes better than previous metrics across various image distributions and is responsive to algorithmic improvements of text-to-image generative models, making it a preferable evaluation metric for these models. We also investigate the design of the evaluation prompts for text-to-image generative models, to make the evaluation stable, fair and easy-to-use. Finally, we establish a benchmark for text-to-image generative models using HPS v2, which includes a set of recent text-to-image models from the academia, community and industry. The code and dataset is / will be available at https://github.com/tgxs002/HPSv2.

Evaluating text-to-image generation models requires alignment with human perception, yet existing human-centric metrics are constrained by limited data coverage, suboptimal feature extraction, and inefficient loss functions. To address these challenges, we introduce Human Preference Score v3 (HPSv3). (1) We release HPDv3, the first wide-spectrum human preference dataset integrating 1.08M text-image pairs and 1.17M annotated pairwise comparisons from state-of-the-art generative models and low to high-quality real-world images. (2) We introduce a VLM-based preference model trained using an uncertainty-aware ranking loss for fine-grained ranking. Besides, we propose Chain-of-Human-Preference (CoHP), an iterative image refinement method that enhances quality without extra data, using HPSv3 to select the best image at each step. Extensive experiments demonstrate that HPSv3 serves as a robust metric for wide-spectrum image evaluation, and CoHP offers an efficient and human-aligned approach to improve image generation quality. The code and dataset are available at the HPSv3 Homepage.

Automatic medical report generation supports clinical diagnosis, reduces the workload of radiologists, and holds the promise of improving diagnosis consistency. However, existing evaluation metrics primarily assess the accuracy of key medical information coverage in generated reports compared to human-written reports, while overlooking crucial details such as the location and certainty of reported abnormalities. These limitations hinder the comprehensive assessment of the reliability of generated reports and pose risks in their selection for clinical use. Therefore, we propose a Granular Explainable Multi-Agent Score (GEMA-Score) in this paper, which conducts both objective quantification and subjective evaluation through a large language model-based multi-agent workflow. Our GEMA-Score parses structured reports and employs NER-F1 calculations through interactive exchanges of information among agents to assess disease diagnosis, location, severity, and uncertainty. Additionally, an LLM-based scoring agent evaluates completeness, readability, and clinical terminology while providing explanatory feedback. Extensive experiments validate that GEMA-Score achieves the highest correlation with human expert evaluations on a public dataset, demonstrating its effectiveness in clinical scoring (Kendall coefficient = 0.70 for Rexval dataset and Kendall coefficient = 0.54 for RadEvalX dataset). The anonymous project demo is available at: https://github.com/Zhenxuan-Zhang/GEMA_score.

Recent advances in text-to-video generation have produced increasingly realistic and diverse content, yet evaluating such videos remains a fundamental challenge due to their multi-faceted nature encompassing visual quality, semantic alignment, and physical consistency. Existing evaluators and reward models are limited to single opaque scores, lack interpretability, or provide only coarse analysis, making them insufficient for capturing the comprehensive nature of video quality assessment. We present VideoScore2, a multi-dimensional, interpretable, and human-aligned framework that explicitly evaluates visual quality, text-to-video alignment, and physical/common-sense consistency while producing detailed chain-of-thought rationales. Our model is trained on a large-scale dataset VideoFeedback2 containing 27,168 human-annotated videos with both scores and reasoning traces across three dimensions, using a two-stage pipeline of supervised fine-tuning followed by reinforcement learning with Group Relative Policy Optimization (GRPO) to enhance analytical robustness. Extensive experiments demonstrate that VideoScore2 achieves superior performance with 44.35 (+5.94) accuracy on our in-domain benchmark VideoScore-Bench-v2 and 50.37 (+4.32) average performance across four out-of-domain benchmarks (VideoGenReward-Bench, VideoPhy2, etc), while providing interpretable assessments that bridge the gap between evaluation and controllable generation through effective reward modeling for Best-of-N sampling. Project Page: https://tiger-ai-lab.github.io/VideoScore2/

Recent advances in instruction-guided image editing underscore the need for effective automated evaluation. While Vision-Language Models (VLMs) have been explored as judges, open-source models struggle with alignment, and proprietary models lack transparency and cost efficiency. Additionally, no public training datasets exist to fine-tune open-source VLMs, only small benchmarks with diverse evaluation schemes. To address this, we introduce ADIEE, an automated dataset creation approach which is then used to train a scoring model for instruction-guided image editing evaluation. We generate a large-scale dataset with over 100K samples and use it to fine-tune a LLaVA-NeXT-8B model modified to decode a numeric score from a custom token. The resulting scorer outperforms all open-source VLMs and Gemini-Pro 1.5 across all benchmarks, achieving a 0.0696 (+17.24%) gain in score correlation with human ratings on AURORA-Bench, and improving pair-wise comparison accuracy by 4.03% (+7.21%) on GenAI-Bench and 4.75% (+9.35%) on AURORA-Bench, respectively, compared to the state-of-the-art. The scorer can act as a reward model, enabling automated best edit selection and model fine-tuning. Notably, the proposed scorer can boost MagicBrush model's average evaluation score on ImagenHub from 5.90 to 6.43 (+8.98%). Our code and models are available at https://github.com/SherryXTChen/ADIEE.git.

Recent advancements in audio-visual generative modeling have been propelled by progress in deep learning and the availability of data-rich benchmarks. However, the growth is not attributed solely to models and benchmarks. Universally accepted evaluation metrics also play an important role in advancing the field. While there are many metrics available to evaluate audio and visual content separately, there is a lack of metrics that offer a quantitative and interpretable measure of audio-visual synchronization for videos "in the wild". To address this gap, we first created a large scale human annotated dataset (100+ hrs) representing nine types of synchronization errors in audio-visual content and how human perceive them. We then developed a PEAVS (Perceptual Evaluation of Audio-Visual Synchrony) score, a novel automatic metric with a 5-point scale that evaluates the quality of audio-visual synchronization. We validate PEAVS using a newly generated dataset, achieving a Pearson correlation of 0.79 at the set level and 0.54 at the clip level when compared to human labels. In our experiments, we observe a relative gain 50% over a natural extension of Fr\'echet based metrics for Audio-Visual synchrony, confirming PEAVS efficacy in objectively modeling subjective perceptions of audio-visual synchronization for videos "in the wild".

Rigorously evaluating machine intelligence against the broad spectrum of human general intelligence has become increasingly important and challenging in this era of rapid technological advance. Conventional AI benchmarks typically assess only narrow capabilities in a limited range of human activity. Most are also static, quickly saturating as developers explicitly or implicitly optimize for them. We propose that a more promising way to evaluate human-like general intelligence in AI systems is through a particularly strong form of general game playing: studying how and how well they play and learn to play all conceivable human games, in comparison to human players with the same level of experience, time, or other resources. We define a "human game" to be a game designed by humans for humans, and argue for the evaluative suitability of this space of all such games people can imagine and enjoy -- the "Multiverse of Human Games". Taking a first step towards this vision, we introduce the AI GameStore, a scalable and open-ended platform that uses LLMs with humans-in-the-loop to synthesize new representative human games, by automatically sourcing and adapting standardized and containerized variants of game environments from popular human digital gaming platforms. As a proof of concept, we generated 100 such games based on the top charts of Apple App Store and Steam, and evaluated seven frontier vision-language models (VLMs) on short episodes of play. The best models achieved less than 10\% of the human average score on the majority of the games, and especially struggled with games that challenge world-model learning, memory and planning. We conclude with a set of next steps for building out the AI GameStore as a practical way to measure and drive progress toward human-like general intelligence in machines.

As interest grows in generating long, detailed image captions, standard evaluation metrics become increasingly unreliable. N-gram-based metrics though efficient, fail to capture semantic correctness. Representational Similarity (RS) metrics, designed to address this, initially saw limited use due to high computational costs, while today, despite advances in hardware, they remain unpopular due to low correlation to human judgments. Meanwhile, metrics based on large language models (LLMs) show strong correlation with human judgments, but remain too expensive for iterative use during model development. We introduce SPECS (Specificity-Enhanced CLIPScore), a reference-free RS metric tailored to long image captioning. SPECS modifies CLIP with a new objective that emphasizes specificity: rewarding correct details and penalizing incorrect ones. We show that SPECS matches the performance of open-source LLM-based metrics in correlation to human judgments, while being far more efficient. This makes it a practical alternative for iterative checkpoint evaluation during image captioning model development.Our code can be found at https://github.com/mbzuai-nlp/SPECS.

One of the most widely used tasks to evaluate Large Language Models (LLMs) is Multiple-Choice Question Answering (MCQA). While open-ended question answering tasks are more challenging to evaluate, MCQA tasks are, in principle, easier to assess, as the model's answer is thought to be simple to extract and is directly compared to a set of predefined choices. However, recent studies have started to question the reliability of MCQA evaluation, showing that multiple factors can significantly impact the reported performance of LLMs, especially when the model generates free-form text before selecting one of the answer choices. In this work, we shed light on the inconsistencies of MCQA evaluation strategies, which can lead to inaccurate and misleading model comparisons. We systematically analyze whether existing answer extraction methods are aligned with human judgment, and how they are influenced by answer constraints in the prompt across different domains. Our experiments demonstrate that traditional evaluation strategies often underestimate LLM capabilities, while LLM-based answer extractors are prone to systematic errors. Moreover, we reveal a fundamental trade-off between including format constraints in the prompt to simplify answer extraction and allowing models to generate free-form text to improve reasoning. Our findings call for standardized evaluation methodologies and highlight the need for more reliable and consistent MCQA evaluation practices.

A trustworthy real-world prediction system should produce well-calibrated confidence scores; that is, its confidence in an answer should be indicative of the likelihood that the answer is correct, enabling deferral to an expert in cases of low-confidence predictions. Recent studies have shown that unsupervised pre-training produces large language models (LMs) whose conditional probabilities are remarkably well-calibrated. However, the most widely-used LMs are fine-tuned with reinforcement learning from human feedback (RLHF-LMs), and some studies have suggested that RLHF-LMs produce conditional probabilities that are very poorly calibrated. In light of this perceived weakness, we conduct a broad evaluation of methods for extracting confidence scores from RLHF-LMs. For RLHF-LMs such as ChatGPT, GPT-4, and Claude, we find that verbalized confidences emitted as output tokens are typically better-calibrated than the model's conditional probabilities on the TriviaQA, SciQ, and TruthfulQA benchmarks, often reducing the expected calibration error by a relative 50%.

Recently, Text-to-Image (T2I) generation models have achieved significant advancements. Correspondingly, many automated metrics have emerged to evaluate the image-text alignment capabilities of generative models. However, the performance comparison among these automated metrics is limited by existing small datasets. Additionally, these datasets lack the capacity to assess the performance of automated metrics at a fine-grained level. In this study, we contribute an EvalMuse-40K benchmark, gathering 40K image-text pairs with fine-grained human annotations for image-text alignment-related tasks. In the construction process, we employ various strategies such as balanced prompt sampling and data re-annotation to ensure the diversity and reliability of our benchmark. This allows us to comprehensively evaluate the effectiveness of image-text alignment metrics for T2I models. Meanwhile, we introduce two new methods to evaluate the image-text alignment capabilities of T2I models: FGA-BLIP2 which involves end-to-end fine-tuning of a vision-language model to produce fine-grained image-text alignment scores and PN-VQA which adopts a novel positive-negative VQA manner in VQA models for zero-shot fine-grained evaluation. Both methods achieve impressive performance in image-text alignment evaluations. We also use our methods to rank current AIGC models, in which the results can serve as a reference source for future study and promote the development of T2I generation. The data and code will be made publicly available.

We introduce DAHL, a benchmark dataset and automated evaluation system designed to assess hallucination in long-form text generation, specifically within the biomedical domain. Our benchmark dataset, meticulously curated from biomedical research papers, consists of 8,573 questions across 29 categories. DAHL evaluates fact-conflicting hallucinations in Large Language Models (LLMs) by deconstructing responses into atomic units, each representing a single piece of information. The accuracy of these responses is averaged to produce the DAHL Score, offering a more in-depth evaluation of hallucinations compared to previous methods that rely on multiple-choice tasks. We conduct experiments with 8 different models, finding that larger models tend to hallucinate less; however, beyond a model size of 7 to 8 billion parameters, further scaling does not significantly improve factual accuracy. The DAHL Score holds potential as an efficient alternative to human-annotated preference labels, being able to be expanded to other specialized domains. We release the dataset and code in public.

Recent advancements in text-to-image (T2I) generation models have transformed the field. However, challenges persist in generating images that reflect demanding textual descriptions, especially for fine-grained details and unusual relationships. Existing evaluation metrics focus on text-image alignment but overlook the realism of the generated image, which can be crucial for downstream applications like data augmentation in machine learning. To address this gap, we propose REAL, an automatic evaluation framework that assesses realism of T2I outputs along three dimensions: fine-grained visual attributes, unusual visual relationships, and visual styles. REAL achieves a Spearman's rho score of up to 0.62 in alignment with human judgement and demonstrates utility in ranking and filtering augmented data for tasks like image captioning, classification, and visual relationship detection. Empirical results show that high-scoring images evaluated by our metrics improve F1 scores of image classification by up to 11.3%, while low-scoring ones degrade that by up to 4.95%. We benchmark four major T2I models across the realism dimensions, providing insights for future improvements in T2I output realism.

Multimodal video captioning condenses dense footage into a structured format of keyframes and natural language. By creating a cohesive multimodal summary, this approach anchors generative AI in rich semantic evidence and serves as a lightweight proxy for high-efficiency retrieval. However, traditional metrics like BLEU or ROUGE fail to quantify information coverage across disparate modalities, such as comparing a paragraph of text to a sequence of keyframes. To address this, we propose the Video Summary Information Loss (ViSIL) score, an information-theoretic framework that quantifies the video information not captured by a summary via vision-language model (VLM) inference. By measuring the information loss, ViSIL is a unified metric that enables direct comparison across multimodal summary formats despite their structural discrepancies. Our results demonstrate that ViSIL scores show a statistically significant correlation with both human and VLM performance on Video Question Answering (VQA) tasks. ViSIL also enables summary selection to optimize the trade-off between information loss and processing speed, establishing a Pareto-optimal frontier that outperforms text summaries by 7% in VQA accuracy without increasing processing load.

Recently, using a powerful proprietary Large Language Model (LLM) (e.g., GPT-4) as an evaluator for long-form responses has become the de facto standard. However, for practitioners with large-scale evaluation tasks and custom criteria in consideration (e.g., child-readability), using proprietary LLMs as an evaluator is unreliable due to the closed-source nature, uncontrolled versioning, and prohibitive costs. In this work, we propose Prometheus, a fully open-source LLM that is on par with GPT-4's evaluation capabilities when the appropriate reference materials (reference answer, score rubric) are accompanied. We first construct the Feedback Collection, a new dataset that consists of 1K fine-grained score rubrics, 20K instructions, and 100K responses and language feedback generated by GPT-4. Using the Feedback Collection, we train Prometheus, a 13B evaluator LLM that can assess any given long-form text based on customized score rubric provided by the user. Experimental results show that Prometheus scores a Pearson correlation of 0.897 with human evaluators when evaluating with 45 customized score rubrics, which is on par with GPT-4 (0.882), and greatly outperforms ChatGPT (0.392). Furthermore, measuring correlation with GPT-4 with 1222 customized score rubrics across four benchmarks (MT Bench, Vicuna Bench, Feedback Bench, Flask Eval) shows similar trends, bolstering Prometheus's capability as an evaluator LLM. Lastly, Prometheus achieves the highest accuracy on two human preference benchmarks (HHH Alignment & MT Bench Human Judgment) compared to open-sourced reward models explicitly trained on human preference datasets, highlighting its potential as an universal reward model. We open-source our code, dataset, and model at https://github.com/kaistAI/Prometheus.

Many recently published Text-to-Speech (TTS) systems produce audio close to real speech. However, TTS evaluation needs to be revisited to make sense of the results obtained with the new architectures, approaches and datasets. We propose evaluating the quality of synthetic speech as a combination of multiple factors such as prosody, speaker identity, and intelligibility. Our approach assesses how well synthetic speech mirrors real speech by obtaining correlates of each factor and measuring their distance from both real speech datasets and noise datasets. We benchmark 35 TTS systems developed between 2008 and 2024 and show that our score computed as an unweighted average of factors strongly correlates with the human evaluations from each time period.

Recent advances in large language models (LLMs) have significantly improved text-to-speech (TTS) systems, enhancing control over speech style, naturalness, and emotional expression, which brings TTS Systems closer to human-level performance. Although the Mean Opinion Score (MOS) remains the standard for TTS System evaluation, it suffers from subjectivity, environmental inconsistencies, and limited interpretability. Existing evaluation datasets also lack a multi-dimensional design, often neglecting factors such as speaking styles, context diversity, and trap utterances, which is particularly evident in Chinese TTS evaluation. To address these challenges, we introduce the Audio Turing Test (ATT), a multi-dimensional Chinese corpus dataset ATT-Corpus paired with a simple, Turing-Test-inspired evaluation protocol. Instead of relying on complex MOS scales or direct model comparisons, ATT asks evaluators to judge whether a voice sounds human. This simplification reduces rating bias and improves evaluation robustness. To further support rapid model development, we also finetune Qwen2-Audio-Instruct with human judgment data as Auto-ATT for automatic evaluation. Experimental results show that ATT effectively differentiates models across specific capability dimensions using its multi-dimensional design. Auto-ATT also demonstrates strong alignment with human evaluations, confirming its value as a fast and reliable assessment tool. The white-box ATT-Corpus and Auto-ATT can be found in ATT Hugging Face Collection (https://huggingface.co/collections/meituan/audio-turing-test-682446320368164faeaf38a4).

Large Language Models are prompting us to view more NLP tasks from a generative perspective. At the same time, they offer a new way of accessing information, mainly through the RAG framework. While there have been notable improvements for the autoregressive models, overcoming hallucination in the generated answers remains a continuous problem. A standard solution is to use commercial LLMs, such as GPT4, to evaluate these algorithms. However, such frameworks are expensive and not very transparent. Therefore, we propose a study which demonstrates the interest of open-weight models for evaluating RAG hallucination. We develop a lightweight approach using smaller, quantized LLMs to provide an accessible and interpretable metric that gives continuous scores for the generated answer with respect to their correctness and faithfulness. This score allows us to question decisions' reliability and explore thresholds to develop a new AUC metric as an alternative to correlation with human judgment.

We present TMMLU+, a new benchmark designed for Traditional Chinese language understanding. TMMLU+ is a multi-choice question-answering dataset with 66 subjects from elementary to professional level. It is six times larger and boasts a more balanced subject distribution than its predecessor, Taiwan Massive Multitask Language Understanding (TMMLU). We also benchmark closed-source models and 26 open-weight Chinese large language models (LLMs) of parameters ranging from 1.8B to 72B on the proposed TMMLU+. Our findings reveal that (1.) Traditional Chinese models still trail behind their Simplified Chinese counterparts, highlighting a need for more focused advancements in LLMs catering to Traditional Chinese. (2.) Current LLMs still fall short of human performance in average scores, indicating a potential need for future research to delve deeper into social science and humanities subjects. (3.) Among all the tokenization compression metrics examined, we identify that only the fertility score uniquely demonstrates strong correlations with our benchmark results. We foresee that TMMLU+ will pinpoint areas for future model improvement, thereby narrowing the gap between machine and human linguistic capabilities and supporting researchers in developing Traditional Chinese LLMs. Our dataset, along with the benchmark source code, is accessible at huggingface.co/datasets/ikala/tmmluplus.

While dialogue remains an important end-goal of natural language research, the difficulty of evaluation is an oft-quoted reason why it remains troublesome to make real progress towards its solution. Evaluation difficulties are actually two-fold: not only do automatic metrics not correlate well with human judgments, but also human judgments themselves are in fact difficult to measure. The two most used human judgment tests, single-turn pairwise evaluation and multi-turn Likert scores, both have serious flaws as we discuss in this work. We instead provide a novel procedure involving comparing two full dialogues, where a human judge is asked to pay attention to only one speaker within each, and make a pairwise judgment. The questions themselves are optimized to maximize the robustness of judgments across different annotators, resulting in better tests. We also show how these tests work in self-play model chat setups, resulting in faster, cheaper tests. We hope these tests become the de facto standard, and will release open-source code to that end.

Automatically evaluating the quality of language generation is critical. Although recent learned metrics show high correlation with human judgement, these metrics can not explain their verdict or associate the scores with defects in generated text. To address this limitation, we present InstructScore, an explainable evaluation metric for text generation. By harnessing both explicit human instruction and the implicit knowledge of GPT-4, we fine-tune a text evaluation metric based on LLaMA, producing both a score for generated text and a human readable diagnostic report. We evaluate InstructScore on a variety of generation tasks, including translation, captioning, data-to-text and commonsense generation. Experiments show that our 7B model surpasses all other unsupervised metrics, including those based on 175B GPT-3 and GPT-4. Surprisingly, our InstructScore, even without direct supervision from human-rated data, achieves performance levels on par with state-of-the-art metrics like COMET22, which were fine-tuned on human ratings.

Precisely evaluating semantic alignment between text prompts and generated videos remains a challenge in Text-to-Video (T2V) Generation. Existing text-to-video alignment metrics like CLIPScore only generate coarse-grained scores without fine-grained alignment details, failing to align with human preference. To address this limitation, we propose ETVA, a novel Evaluation method of Text-to-Video Alignment via fine-grained question generation and answering. First, a multi-agent system parses prompts into semantic scene graphs to generate atomic questions. Then we design a knowledge-augmented multi-stage reasoning framework for question answering, where an auxiliary LLM first retrieves relevant common-sense knowledge (e.g., physical laws), and then video LLM answers the generated questions through a multi-stage reasoning mechanism. Extensive experiments demonstrate that ETVA achieves a Spearman's correlation coefficient of 58.47, showing a much higher correlation with human judgment than existing metrics which attain only 31.0. We also construct a comprehensive benchmark specifically designed for text-to-video alignment evaluation, featuring 2k diverse prompts and 12k atomic questions spanning 10 categories. Through a systematic evaluation of 15 existing text-to-video models, we identify their key capabilities and limitations, paving the way for next-generation T2V generation.

Evaluating machine-generated text remains a significant challenge in NLP, especially for non-English languages. Current methodologies, including automated metrics, human assessments, and LLM-based evaluations, predominantly focus on English, revealing a significant gap in multilingual evaluation frameworks. We introduce the Cross Lingual Auto Evaluation (CIA) Suite, an extensible framework that includes evaluator LLMs (Hercule) and a novel test set (Recon) specifically designed for multilingual evaluation. Our test set features 500 human-annotated instructions spanning various task capabilities along with human judgment scores across six languages. This would enable benchmarking of general-purpose multilingual LLMs and facilitate meta-evaluation of Evaluator LLMs. The proposed model, Hercule, is a cross-lingual evaluation model that addresses the scarcity of reference answers in the target language by learning to assign scores to responses based on easily available reference answers in English. Our experiments demonstrate that Hercule aligns more closely with human judgments compared to proprietary models, demonstrating the effectiveness of such cross-lingual evaluation in low resource scenarios. Further, it is also effective in zero-shot evaluation on unseen languages. This study is the first comprehensive examination of cross-lingual evaluation using LLMs, presenting a scalable and effective approach for multilingual assessment. All code, datasets, and models will be publicly available to enable further research in this important area.

We conduct a systematic audit of three widely used reasoning benchmarks, SocialIQa, FauxPas-EAI, and ToMi, and uncover pervasive flaws in both benchmark items and evaluation methodology. Using five LLMs (GPT-{3, 3.5, 4, o1}, and LLaMA 3.1) as diagnostic tools, we identify structural, semantic, and pragmatic issues in benchmark design (e.g., duplicated items, ambiguous wording, and implausible answers), as well as scoring procedures that prioritize output form over reasoning process. Through systematic human annotation and re-evaluation on cleaned benchmark subsets, we find that model scores often improve not due to due to erratic surface wording variations and not to improved reasoning. Infact, further analyses show that model performance is highly sensitive to minor input variations such as context availability and phrasing, revealing that high scores may reflect alignment with format-specific cues rather than consistent inference based on the input. These findings challenge the validity of current benchmark-based claims about reasoning in LLMs, and highlight the need for evaluation protocols that assess reasoning as a process of drawing inference from available information, rather than as static output selection. We release audited data and evaluation tools to support more interpretable and diagnostic assessments of model reasoning.

Battles, or side-by-side comparisons in so called arenas that elicit human preferences, have emerged as a popular approach to assessing the output quality of LLMs. Recently, this idea has been extended to retrieval-augmented generation (RAG) systems. While undoubtedly representing an advance in evaluation, battles have at least two drawbacks, particularly in the context of complex information-seeking queries: they are neither explanatory nor diagnostic. Recently, the nugget evaluation methodology has emerged as a promising approach to evaluate the quality of RAG answers. Nuggets decompose long-form LLM-generated answers into atomic facts, highlighting important pieces of information necessary in a "good" response. In this work, we apply our AutoNuggetizer framework to analyze data from roughly 7K Search Arena battles provided by LMArena in a fully automatic manner. Our results show a significant correlation between nugget scores and human preferences, showcasing promise in our approach to explainable and diagnostic system evaluations.

Trainable evaluation metrics for machine translation (MT) exhibit strong correlation with human judgements, but they are often hard to interpret and might produce unreliable scores under noisy or out-of-domain data. Recent work has attempted to mitigate this with simple uncertainty quantification techniques (Monte Carlo dropout and deep ensembles), however these techniques (as we show) are limited in several ways -- for example, they are unable to distinguish between different kinds of uncertainty, and they are time and memory consuming. In this paper, we propose more powerful and efficient uncertainty predictors for MT evaluation, and we assess their ability to target different sources of aleatoric and epistemic uncertainty. To this end, we develop and compare training objectives for the COMET metric to enhance it with an uncertainty prediction output, including heteroscedastic regression, divergence minimization, and direct uncertainty prediction. Our experiments show improved results on uncertainty prediction for the WMT metrics task datasets, with a substantial reduction in computational costs. Moreover, they demonstrate the ability of these predictors to address specific uncertainty causes in MT evaluation, such as low quality references and out-of-domain data.

Before deploying a language model (LM) within a given domain, it is important to measure its tendency to generate factually incorrect information in that domain. Existing factual generation evaluation methods focus on facts sampled from the LM itself, and thus do not control the set of evaluated facts and might under-represent rare and unlikely facts. We propose FACTOR: Factual Assessment via Corpus TransfORmation, a scalable approach for evaluating LM factuality. FACTOR automatically transforms a factual corpus of interest into a benchmark evaluating an LM's propensity to generate true facts from the corpus vs. similar but incorrect statements. We use our framework to create two benchmarks: Wiki-FACTOR and News-FACTOR. We show that: (i) our benchmark scores increase with model size and improve when the LM is augmented with retrieval; (ii) benchmark score correlates with perplexity, but the two metrics do not always agree on model ranking; and (iii) when perplexity and benchmark score disagree, the latter better reflects factuality in open-ended generation, as measured by human annotators. We make our data and code publicly available in https://github.com/AI21Labs/factor.

Comparing human and model performance offers a valuable perspective for understanding the strengths and limitations of embedding models, highlighting where they succeed and where they fail to capture meaning and nuance. However, such comparisons are rarely made, as human performance on embedding tasks is difficult to measure. To fill this gap, we introduce HUME: Human Evaluation Framework for Text Embeddings. While frameworks like MTEB provide broad model evaluation, they lack reliable estimates of human performance, limiting the interpretability of model scores. We measure human performance across 16 MTEB datasets spanning reranking, classification, clustering, and semantic textual similarity across linguistically diverse high- and low-resource languages. Humans achieve an average performance of 77.6% compared to 80.1% for the best embedding model, although variation is substantial: models reach near-ceiling performance on some datasets while struggling on others, suggesting dataset issues and revealing shortcomings in low-resource languages. We provide human performance baselines, insight into task difficulty patterns, and an extensible evaluation framework that enables a more meaningful interpretation of the model and informs the development of both models and benchmarks. Our code, dataset, and leaderboard are publicly available at https://github.com/embeddings-benchmark/mteb.

Large Language Models (LLMs) have shown promise in accelerating the scientific research pipeline. A key capability for this process is the ability to generate novel research ideas, and prior studies have found settings in which LLM-generated research ideas were judged as more novel than human-expert ideas. However, a good idea should not simply appear to be novel, it should also result in better research after being executed. To test whether AI-generated ideas lead to better research outcomes, we conduct an execution study by recruiting 43 expert researchers to execute randomly-assigned ideas, either written by experts or generated by an LLM. Each expert spent over 100 hours implementing the idea and wrote a 4-page short paper to document the experiments. All the executed projects are then reviewed blindly by expert NLP researchers. Comparing the review scores of the same ideas before and after execution, the scores of the LLM-generated ideas decrease significantly more than expert-written ideas on all evaluation metrics (novelty, excitement, effectiveness, and overall; p < 0.05), closing the gap between LLM and human ideas observed at the ideation stage. When comparing the aggregated review scores from the execution study, we even observe that for many metrics there is a flip in rankings where human ideas score higher than LLM ideas. This ideation-execution gap highlights the limitations of current LLMs in generating truly effective research ideas and the challenge of evaluating research ideas in the absence of execution outcomes.

Evaluating large language models (LLMs) for instrument control requires methods that go beyond standard, stateless algorithmic benchmarks, since the behavior of physical systems cannot be fully captured by unit tests alone. Here we introduce EnvTrace, a simulation-based method that evaluates execution traces to assess semantic code equivalence. EnvTrace is demonstrated with a beamline control-logic digital twin to facilitate the evaluation of instrument control code, with the digital twin itself also enabling the pre-execution validation of live experiments. Over 30 LLMs were evaluated using trace alignment to generate a multi-faceted score for functional correctness across key behavioral dimensions, showing that many top-tier models can approach human-level performance in rapid control-code generation. This is a first step toward a broader vision where LLMs and digital twins work symbiotically: LLMs providing intuitive control and agentic orchestration, and digital twins offering safe and high-fidelity environments, paving the way towards autonomous embodied AI.

Existing automatic evaluation on text-to-image synthesis can only provide an image-text matching score, without considering the object-level compositionality, which results in poor correlation with human judgments. In this work, we propose LLMScore, a new framework that offers evaluation scores with multi-granularity compositionality. LLMScore leverages the large language models (LLMs) to evaluate text-to-image models. Initially, it transforms the image into image-level and object-level visual descriptions. Then an evaluation instruction is fed into the LLMs to measure the alignment between the synthesized image and the text, ultimately generating a score accompanied by a rationale. Our substantial analysis reveals the highest correlation of LLMScore with human judgments on a wide range of datasets (Attribute Binding Contrast, Concept Conjunction, MSCOCO, DrawBench, PaintSkills). Notably, our LLMScore achieves Kendall's tau correlation with human evaluations that is 58.8% and 31.2% higher than the commonly-used text-image matching metrics CLIP and BLIP, respectively.

Despite the recent advances in Large Language Models, benchmarks for evaluating legal writing remain scarce due to the inherent complexity of assessing open-ended responses in this domain. One of the key challenges in evaluating language models on domain-specific tasks is finding test datasets that are public, frequently updated, and contain comprehensive evaluation guidelines. The Brazilian Bar Examination meets these requirements. We introduce oab-bench, a benchmark comprising 105 questions across seven areas of law from recent editions of the exam. The benchmark includes comprehensive evaluation guidelines and reference materials used by human examiners to ensure consistent grading. We evaluate the performance of four LLMs on oab-bench, finding that Claude-3.5 Sonnet achieves the best results with an average score of 7.93 out of 10, passing all 21 exams. We also investigated whether LLMs can serve as reliable automated judges for evaluating legal writing. Our experiments show that frontier models like OpenAI's o1 achieve a strong correlation with human scores when evaluating approved exams, suggesting their potential as reliable automated evaluators despite the inherently subjective nature of legal writing assessment. The source code and the benchmark -- containing questions, evaluation guidelines, model-generated responses, and their respective automated evaluations -- are publicly available.

Studies on evaluation metrics and LLM-as-a-Judge models for automatic text summarization have largely been focused on English, limiting our understanding of their effectiveness in other languages. Through our new dataset BASSE (BAsque and Spanish Summarization Evaluation), we address this situation by collecting human judgments on 2,040 abstractive summaries in Basque and Spanish, generated either manually or by five LLMs with four different prompts. For each summary, annotators evaluated five criteria on a 5-point Likert scale: coherence, consistency, fluency, relevance, and 5W1H. We use these data to reevaluate traditional automatic metrics used for evaluating summaries, as well as several LLM-as-a-Judge models that show strong performance on this task in English. Our results show that currently proprietary judge LLMs have the highest correlation with human judgments, followed by criteria-specific automatic metrics, while open-sourced judge LLMs perform poorly. We release BASSE and our code publicly, along with the first large-scale Basque summarization dataset containing 22,525 news articles with their subheads.

Reinforcement learning from human feedback (RLHF) has emerged as a powerful technique to make large language models (LLMs) more capable in complex settings. RLHF proceeds as collecting human preference data, training a reward model on said data, and optimizing a base ML model with respect to said reward for extrinsic evaluation metrics (e.g. MMLU, GSM8k). RLHF relies on many assumptions about how the various pieces fit together, such as a reward model capturing human preferences and an RL optimizer extracting the right signal from a reward model. As the RLHF process involves many distinct design decisions, it is easy to assume that multiple processes are correlated and therefore numerically linked. This apparent correlation is often not true, where reward models are easily overoptimized or RL optimizers can reduce performance on tasks not modeled in the data. Notable manifestations of models trained with imperfect RLHF systems are those that are prone to refusing basic requests for safety reasons or appearing lazy in generations. As chat model evaluation becomes increasingly nuanced, the reliance on a perceived link between reward model training, RL scores, and downstream performance drives these issues, which we describe as an objective mismatch. In this paper, we illustrate the causes of this issue, reviewing relevant literature from model-based reinforcement learning, and argue for solutions. By solving objective mismatch in RLHF, the ML models of the future will be more precisely aligned to user instructions for both safety and helpfulness.

Distributional metrics such as Fréchet Audio Distance cannot score individual music clips and correlate poorly with human judgments, while the only per-sample learned metric achieving high human correlation is closed-source. We introduce MUQ-EVAL, an open-source per-sample quality metric for AIgenerated music built by training lightweight prediction heads on frozen MuQ-310M features using MusicEval, a dataset of generated clips from 31 text-to-music systems with expert quality ratings. Our simplest model, frozen features with attention pooling and a two-layer MLP, achieves system-level SRCC = 0.957 and utterance-level SRCC = 0.838 with human mean opinion scores. A systematic ablation over training objectives and adaptation strategies shows that no addition meaningfully improves the frozen baseline, indicating that frozen MuQ representations already capture quality-relevant information. Encoder choice is the dominant design factor, outweighing all architectural and training decisions. LoRA-adapted models trained on as few as 150 clips already achieve usable correlation, enabling personalized quality evaluators from individual listener annotations. A controlled degradation analysis reveals selective sensitivity to signal-level artifacts but insensitivity to musical-structural distortions. Our metric, MUQ-EVAL, is fully open-source, outperforms existing open per-sample metrics, and runs in real time on a single consumer GPU. Code, model weights, and evaluation scripts are available at https://github.com/dgtql/MuQ-Eval.

Evaluating Large Language Models (LLMs) for mental health support is challenging due to the emotionally and cognitively complex nature of therapeutic dialogue. Existing benchmarks are limited in scale, reliability, often relying on synthetic or social media data, and lack frameworks to assess when automated judges can be trusted. To address the need for large-scale dialogue datasets and judge reliability assessment, we introduce two benchmarks that provide a framework for generation and evaluation. MentalBench-100k consolidates 10,000 one-turn conversations from three real scenarios datasets, each paired with nine LLM-generated responses, yielding 100,000 response pairs. MentalAlign-70k}reframes evaluation by comparing four high-performing LLM judges with human experts across 70,000 ratings on seven attributes, grouped into Cognitive Support Score (CSS) and Affective Resonance Score (ARS). We then employ the Affective Cognitive Agreement Framework, a statistical methodology using intraclass correlation coefficients (ICC) with confidence intervals to quantify agreement, consistency, and bias between LLM judges and human experts. Our analysis reveals systematic inflation by LLM judges, strong reliability for cognitive attributes such as guidance and informativeness, reduced precision for empathy, and some unreliability in safety and relevance. Our contributions establish new methodological and empirical foundations for reliable, large-scale evaluation of LLMs in mental health. We release the benchmarks and codes at: https://github.com/abeerbadawi/MentalBench/

Evaluating radiology reports is a challenging problem as factual correctness is extremely important due to the need for accurate medical communication about medical images. Existing automatic evaluation metrics either suffer from failing to consider factual correctness (e.g., BLEU and ROUGE) or are limited in their interpretability (e.g., F1CheXpert and F1RadGraph). In this paper, we introduce GREEN (Generative Radiology Report Evaluation and Error Notation), a radiology report generation metric that leverages the natural language understanding of language models to identify and explain clinically significant errors in candidate reports, both quantitatively and qualitatively. Compared to current metrics, GREEN offers: 1) a score aligned with expert preferences, 2) human interpretable explanations of clinically significant errors, enabling feedback loops with end-users, and 3) a lightweight open-source method that reaches the performance of commercial counterparts. We validate our GREEN metric by comparing it to GPT-4, as well as to error counts of 6 experts and preferences of 2 experts. Our method demonstrates not only higher correlation with expert error counts, but simultaneously higher alignment with expert preferences when compared to previous approaches."

Visual grounding, localizing objects from natural language descriptions, represents a critical bridge between language and vision understanding. While multimodal large language models (MLLMs) achieve impressive scores on existing benchmarks, a fundamental question remains: can MLLMs truly ground language in vision with human-like sophistication, or are they merely pattern-matching on simplified datasets? Current benchmarks fail to capture real-world complexity where humans effortlessly navigate ambiguous references and recognize when grounding is impossible. To rigorously assess MLLMs' true capabilities, we introduce GroundingME, a benchmark that systematically challenges models across four critical dimensions: (1) Discriminative, distinguishing highly similar objects, (2) Spatial, understanding complex relational descriptions, (3) Limited, handling occlusions or tiny objects, and (4) Rejection, recognizing ungroundable queries. Through careful curation combining automated generation with human verification, we create 1,005 challenging examples mirroring real-world complexity. Evaluating 25 state-of-the-art MLLMs reveals a profound capability gap: the best model achieves only 45.1% accuracy, while most score 0% on rejection tasks, reflexively hallucinating objects rather than acknowledging their absence, raising critical safety concerns for deployment. We explore two strategies for improvements: (1) test-time scaling selects optimal response by thinking trajectory to improve complex grounding by up to 2.9%, and (2) data-mixture training teaches models to recognize ungroundable queries, boosting rejection accuracy from 0% to 27.9%. GroundingME thus serves as both a diagnostic tool revealing current limitations in MLLMs and a roadmap toward human-level visual grounding.

Multiple choice question answering (MCQA) is popular for LLM evaluation due to its simplicity and human-like testing, but we argue for its reform. We first reveal flaws in MCQA's format, as it struggles to: 1) test generation/subjectivity; 2) match LLM use cases; and 3) fully test knowledge. We instead advocate for generative formats based on human testing-where LLMs construct and explain answers-better capturing user needs and knowledge while remaining easy to score. We then show even when MCQA is a useful format, its datasets suffer from: leakage; unanswerability; shortcuts; and saturation. In each issue, we give fixes from education, like rubrics to guide MCQ writing; scoring methods to bridle guessing; and Item Response Theory to build harder MCQs. Lastly, we discuss LLM errors in MCQA-robustness, biases, and unfaithful explanations-showing how our prior solutions better measure or address these issues. While we do not need to desert MCQA, we encourage more efforts in refining the task based on educational testing, advancing evaluations.

Question answering (QA) can only make progress if we know if an answer is correct, but for many of the most challenging and interesting QA examples, current evaluation metrics to determine answer equivalence (AE) often do not align with human judgments, particularly more verbose, free-form answers from large language models (LLM). There are two challenges: a lack of data and that models are too big: LLM-based scorers can correlate better with human judges, but this task has only been tested on limited QA datasets, and even when available, update of the model is limited because LLMs are large and often expensive. We rectify both of these issues by providing clear and consistent guidelines for evaluating AE in machine QA adopted from professional human QA contests. We also introduce a combination of standard evaluation and a more efficient, robust, and lightweight discriminate AE classifier-based matching method (CFMatch, smaller than 1 MB), trained and validated to more accurately evaluate answer correctness in accordance with adopted expert AE rules that are more aligned with human judgments.

Many decision-making tasks, where both accuracy and efficiency matter, still require human supervision. For example, tasks like traffic officers reviewing hour-long dashcam footage or researchers screening conference videos can benefit from concise summaries that reduce cognitive load and save time. Yet current vision-language models (VLMs) often produce verbose, redundant outputs that hinder task performance. Existing video caption evaluation depends on costly human annotations and overlooks the summaries' utility in downstream tasks. We address these gaps with Video-to-text Information Bottleneck Evaluation (VIBE), an annotation-free method that scores VLM outputs using two metrics: grounding (how well the summary aligns with visual content) and utility (how informative it is for the task). VIBE selects from randomly sampled VLM outputs by ranking them according to the two scores to support effective human decision-making. Human studies on LearningPaper24, SUTD-TrafficQA, and LongVideoBench show that summaries selected by VIBE consistently improve performance-boosting task accuracy by up to 61.23% and reducing response time by 75.77% compared to naive VLM summaries or raw video.

Reward models (RMs) play a crucial role in Reinforcement Learning from Human Feedback by serving as proxies for human preferences in aligning large language models. In this paper, we identify a model preference bias in RMs, where they systematically assign disproportionately high scores to responses from certain policy models. This bias distorts ranking evaluations and leads to unfair judgments. To address this issue, we propose a calibration method named CHatbot Arena calibrated Reward Modeling (CHARM) that leverages Elo scores from the Chatbot Arena leaderboard to mitigate RM overvaluation. We also introduce a Mismatch Degree metric to measure this preference bias. Our approach is computationally efficient, requiring only a small preference dataset for continued training of the RM. We conduct extensive experiments on reward model benchmarks and human preference alignment. Results demonstrate that our calibrated RMs (1) achieve improved evaluation accuracy on RM-Bench and the Chat-Hard domain of RewardBench, and (2) exhibit a stronger correlation with human preferences by producing scores more closely aligned with Elo rankings. By mitigating model preference bias, our method provides a generalizable and efficient solution for building fairer and more reliable reward models.

Scientific data visualization plays a crucial role in research by enabling the direct display of complex information and assisting researchers in identifying implicit patterns. Despite its importance, the use of Large Language Models (LLMs) for scientific data visualization remains rather unexplored. In this study, we introduce MatPlotAgent, an efficient model-agnostic LLM agent framework designed to automate scientific data visualization tasks. Leveraging the capabilities of both code LLMs and multi-modal LLMs, MatPlotAgent consists of three core modules: query understanding, code generation with iterative debugging, and a visual feedback mechanism for error correction. To address the lack of benchmarks in this field, we present MatPlotBench, a high-quality benchmark consisting of 100 human-verified test cases. Additionally, we introduce a scoring approach that utilizes GPT-4V for automatic evaluation. Experimental results demonstrate that MatPlotAgent can improve the performance of various LLMs, including both commercial and open-source models. Furthermore, the proposed evaluation method shows a strong correlation with human-annotated scores.

Video summarization is a technique to create a short skim of the original video while preserving the main stories/content. There exists a substantial interest in automatizing this process due to the rapid growth of the available material. The recent progress has been facilitated by public benchmark datasets, which enable easy and fair comparison of methods. Currently the established evaluation protocol is to compare the generated summary with respect to a set of reference summaries provided by the dataset. In this paper, we will provide in-depth assessment of this pipeline using two popular benchmark datasets. Surprisingly, we observe that randomly generated summaries achieve comparable or better performance to the state-of-the-art. In some cases, the random summaries outperform even the human generated summaries in leave-one-out experiments. Moreover, it turns out that the video segmentation, which is often considered as a fixed pre-processing method, has the most significant impact on the performance measure. Based on our observations, we propose alternative approaches for assessing the importance scores as well as an intuitive visualization of correlation between the estimated scoring and human annotations.

Modern large language models (LLMs) should generally benefit individuals from various cultural backgrounds around the world. However, most recent advanced generative evaluation benchmarks tailed for LLMs mainly focus on English. To this end, we introduce OMGEval, the first Open-source Multilingual Generative test set that can assess the capability of LLMs in different languages. For each language, OMGEval provides 804 open-ended questions, covering a wide range of important capabilities of LLMs, such as general knowledge, logical reasoning, and so on. Each question is rigorously verified by human annotators. Notably, to sufficiently reflect the compatibility of LLMs in different cultural backgrounds, we perform localization for each non-English language. Specifically, the current version of OMGEval includes 5 languages (i.e., Zh, Ru, Fr, Es, Ar). Following AlpacaEval, we employ GPT-4 as the adjudicator to automatically score different model outputs, which is shown closely related to human evaluation. We evaluate several representative multilingual LLMs on the proposed OMGEval, which we believe will provide a valuable reference for the community to further understand and improve the multilingual capability of LLMs. OMGEval is available at https://github.com/blcuicall/OMGEval.

Large Language Model (LLM) personas with explicit specifications of attributes, background, and behavioural tendencies are increasingly used to simulate human conversations for tasks such as user modeling, social reasoning, and behavioural analysis. Ensuring that persona-grounded simulations faithfully reflect human conversational behaviour is therefore critical. However, current evaluation practices largely rely on LLM-as-a-judge approaches, offering limited grounding in observable human behavior and producing opaque scalar scores. We address this gap by proposing Eval4Sim, an evaluation framework that measures how closely simulated conversations align with human conversational patterns across three complementary dimensions. Adherence captures how effectively persona backgrounds are implicitly encoded in generated utterances, assessed via dense retrieval with speaker-aware representations. Consistency evaluates whether a persona maintains a distinguishable identity across conversations, computed through authorship verification. Naturalness reflects whether conversations exhibit human-like flow rather than overly rigid or optimized structure, quantified through distributions derived from dialogue-focused Natural Language Inference. Unlike absolute or optimization-oriented metrics, Eval4Sim uses a human conversational corpus (i.e., PersonaChat) as a reference baseline and penalizes deviations in both directions, distinguishing insufficient persona encoding from over-optimized, unnatural behaviour. Although demonstrated on PersonaChat, the applicability of Eval4Sim extends to any conversational corpus containing speaker-level annotations.

We introduce Vibe-Eval: a new open benchmark and framework for evaluating multimodal chat models. Vibe-Eval consists of 269 visual understanding prompts, including 100 of hard difficulty, complete with gold-standard responses authored by experts. Vibe-Eval is open-ended and challenging with dual objectives: (i) vibe checking multimodal chat models for day-to-day tasks and (ii) rigorously testing and probing the capabilities of present frontier models. Notably, our hard set contains >50% questions that all frontier models answer incorrectly. We explore the nuances of designing, evaluating, and ranking models on ultra challenging prompts. We also discuss trade-offs between human and automatic evaluation, and show that automatic model evaluation using Reka Core roughly correlates to human judgment. We offer free API access for the purpose of lightweight evaluation and plan to conduct formal human evaluations for public models that perform well on the Vibe-Eval's automatic scores. We release the evaluation code and data, see https://github.com/reka-ai/reka-vibe-eval

Automatic dialogue coherence evaluation has attracted increasing attention and is crucial for developing promising dialogue systems. However, existing metrics have two major limitations: (a) they are mostly trained in a simplified two-level setting (coherent vs. incoherent), while humans give Likert-type multi-level coherence scores, dubbed as "quantifiable"; (b) their predicted coherence scores cannot align with the actual human rating standards due to the absence of human guidance during training. To address these limitations, we propose Quantifiable Dialogue Coherence Evaluation (QuantiDCE), a novel framework aiming to train a quantifiable dialogue coherence metric that can reflect the actual human rating standards. Specifically, QuantiDCE includes two training stages, Multi-Level Ranking (MLR) pre-training and Knowledge Distillation (KD) fine-tuning. During MLR pre-training, a new MLR loss is proposed for enabling the model to learn the coarse judgement of coherence degrees. Then, during KD fine-tuning, the pretrained model is further finetuned to learn the actual human rating standards with only very few human-annotated data. To advocate the generalizability even with limited fine-tuning data, a novel KD regularization is introduced to retain the knowledge learned at the pre-training stage. Experimental results show that the model trained by QuantiDCE presents stronger correlations with human judgements than the other state-of-the-art metrics.

Evaluation of Large Language Models (LLMs) is challenging because aligning to human values requires the composition of multiple skills and the required set of skills varies depending on the instruction. Recent studies have evaluated the performance of LLMs in two ways, (1) automatic evaluation on several independent benchmarks and (2) human or machined-based evaluation giving an overall score to the response. However, both settings are coarse-grained evaluations, not considering the nature of user instructions that require instance-wise skill composition, which limits the interpretation of the true capabilities of LLMs. In this paper, we introduce FLASK (Fine-grained Language Model Evaluation based on Alignment SKill Sets), a fine-grained evaluation protocol that can be used for both model-based and human-based evaluation which decomposes coarse-level scoring to an instance-wise skill set-level. Specifically, we define 12 fine-grained skills needed for LLMs to follow open-ended user instructions and construct an evaluation set by allocating a set of skills for each instance. Additionally, by annotating the target domains and difficulty level for each instance, FLASK provides a holistic view with a comprehensive analysis of a model's performance depending on skill, domain, and difficulty. Through using FLASK, we compare multiple open-sourced and proprietary LLMs and observe highly-correlated findings between model-based and human-based evaluations. FLASK enables developers to more accurately measure the model performance and how it can be improved by analyzing factors that make LLMs proficient in particular skills. For practitioners, FLASK can be used to recommend suitable models for particular situations through comprehensive comparison among various LLMs. We release the evaluation data and code implementation at https://github.com/kaistAI/FLASK.

Recent advances in large language models (LLMs) for mathematical reasoning have largely focused on tasks with easily verifiable final answers; however, generating and verifying natural language math proofs remains an open challenge. We identify the absence of a reliable, fine-grained evaluator for LLM-generated math proofs as a critical gap. To address this, we propose a systematic methodology for developing and validating evaluators that assign fine-grained scores on a 0-7 scale to model-generated math proofs. To enable this study, we introduce ProofBench, the first expert-annotated dataset of fine-grained proof ratings, spanning 145 problems from six major math competitions (USAMO, IMO, Putnam, etc) and 435 LLM-generated solutions from Gemini-2.5-pro, o3, and DeepSeek-R1. %with expert gradings. Using ProofBench as a testbed, we systematically explore the evaluator design space across key axes: the backbone model, input context, instructions and evaluation workflow. Our analysis delivers ProofGrader, an evaluator that combines a strong reasoning backbone LM, rich context from reference solutions and marking schemes, and a simple ensembling method; it achieves a low Mean Absolute Error (MAE) of 0.926 against expert scores, significantly outperforming naive baselines. Finally, we demonstrate its practical utility in a best-of-n selection task: at n=16, ProofGrader achieves an average score of 4.14 (out of 7), closing 78% of the gap between a naive binary evaluator (2.48) and the human oracle (4.62), highlighting its potential to advance downstream proof generation.

Machine Translation (MT) evaluation metrics assess translation quality automatically. Recently, researchers have employed MT metrics for various new use cases, such as data filtering and translation re-ranking. However, most MT metrics return assessments as scalar scores that are difficult to interpret, posing a challenge to making informed design choices. Moreover, MT metrics' capabilities have historically been evaluated using correlation with human judgment, which, despite its efficacy, falls short of providing intuitive insights into metric performance, especially in terms of new metric use cases. To address these issues, we introduce an interpretable evaluation framework for MT metrics. Within this framework, we evaluate metrics in two scenarios that serve as proxies for the data filtering and translation re-ranking use cases. Furthermore, by measuring the performance of MT metrics using Precision, Recall, and F-score, we offer clearer insights into their capabilities than correlation with human judgments. Finally, we raise concerns regarding the reliability of manually curated data following the Direct Assessments+Scalar Quality Metrics (DA+SQM) guidelines, reporting a notably low agreement with Multidimensional Quality Metrics (MQM) annotations.

Current metrics for text-to-image models typically rely on statistical metrics which inadequately represent the real preference of humans. Although recent work attempts to learn these preferences via human annotated images, they reduce the rich tapestry of human preference to a single overall score. However, the preference results vary when humans evaluate images with different aspects. Therefore, to learn the multi-dimensional human preferences, we propose the Multi-dimensional Preference Score (MPS), the first multi-dimensional preference scoring model for the evaluation of text-to-image models. The MPS introduces the preference condition module upon CLIP model to learn these diverse preferences. It is trained based on our Multi-dimensional Human Preference (MHP) Dataset, which comprises 918,315 human preference choices across four dimensions (i.e., aesthetics, semantic alignment, detail quality and overall assessment) on 607,541 images. The images are generated by a wide range of latest text-to-image models. The MPS outperforms existing scoring methods across 3 datasets in 4 dimensions, enabling it a promising metric for evaluating and improving text-to-image generation.

As world models gain momentum in Embodied AI, an increasing number of works explore using video foundation models as predictive world models for downstream embodied tasks like 3D prediction or interactive generation. However, before exploring these downstream tasks, video foundation models still have two critical questions unanswered: (1) whether their generative generalization is sufficient to maintain perceptual fidelity in the eyes of human observers, and (2) whether they are robust enough to serve as a universal prior for real-world embodied agents. To provide a standardized framework for answering these questions, we introduce the Embodied Turing Test benchmark: WoW-World-Eval (Wow,wo,val). Building upon 609 robot manipulation data, Wow-wo-val examines five core abilities, including perception, planning, prediction, generalization, and execution. We propose a comprehensive evaluation protocol with 22 metrics to assess the models' generation ability, which achieves a high Pearson Correlation between the overall score and human preference (>0.93) and establishes a reliable foundation for the Human Turing Test. On Wow-wo-val, models achieve only 17.27 on long-horizon planning and at best 68.02 on physical consistency, indicating limited spatiotemporal consistency and physical reasoning. For the Inverse Dynamic Model Turing Test, we first use an IDM to evaluate the video foundation models' execution accuracy in the real world. However, most models collapse to approx 0% success, while WoW maintains a 40.74% success rate. These findings point to a noticeable gap between the generated videos and the real world, highlighting the urgency and necessity of benchmarking World Model in Embodied AI.

We present TransEvalnia, a prompting-based translation evaluation and ranking system that uses reasoning in performing its evaluations and ranking. This system presents fine-grained evaluations based on a subset of the Multidimensional Quality Metrics (https://themqm.org/), returns an assessment of which translation it deems the best, and provides numerical scores for the various dimensions and for the overall translation. We show that TransEvalnia performs as well as or better than the state-of-the-art MT-Ranker (Moosa et al. 2024) on our own English-Japanese data as well as several language pairs from various WMT shared tasks. Using Anthropic's Claude-3.5-Sonnet and Qwen-2.5-72B-Instruct as the evaluation LLMs, we show that the evaluations returned are deemed highly acceptable to human raters, and that the scores assigned to the translations by Sonnet, as well as other LLMs, correlate well with scores assigned by the human raters. We also note the sensitivity of our system -- as well as MT-Ranker -- to the order in which the translations are presented, and we propose methods to address this position bias. All data, including the system's evaluation and reasoning, human assessments, as well as code is released.

Large Language Models (LLMs) are widely used to evaluate natural language generation tasks as automated metrics. However, the likelihood, a measure of LLM's plausibility for a sentence, can vary due to superficial differences in sentences, such as word order and sentence structure. It is therefore possible that there might be a likelihood bias if LLMs are used for evaluation: they might overrate sentences with higher likelihoods while underrating those with lower likelihoods. In this paper, we investigate the presence and impact of likelihood bias in LLM-based evaluators. We also propose a method to mitigate the likelihood bias. Our method utilizes highly biased instances as few-shot examples for in-context learning. Our experiments in evaluating the data-to-text and grammatical error correction tasks reveal that several LLMs we test display a likelihood bias. Furthermore, our proposed method successfully mitigates this bias, also improving evaluation performance (in terms of correlation of models with human scores) significantly.

Providing quality scores along with Machine Translation (MT) output, so-called reference-free Quality Estimation (QE), is crucial to inform users about the reliability of the translation. We propose a model-specific, unsupervised QE approach, termed kNN-QE, that extracts information from the MT model's training data using k-nearest neighbors. Measuring the performance of model-specific QE is not straightforward, since they provide quality scores on their own MT output, thus cannot be evaluated using benchmark QE test sets containing human quality scores on premade MT output. Therefore, we propose an automatic evaluation method that uses quality scores from reference-based metrics as gold standard instead of human-generated ones. We are the first to conduct detailed analyses and conclude that this automatic method is sufficient, and the reference-based MetricX-23 is best for the task.

With the primary focus on evaluating the effectiveness of large language models for automatic reference-less translation assessment, this work presents our experiments on mimicking human direct assessment to evaluate the quality of translations in English and Indian languages. We constructed a translation evaluation task where we performed zero-shot learning, in-context example-driven learning, and fine-tuning of large language models to provide a score out of 100, where 100 represents a perfect translation and 1 represents a poor translation. We compared the performance of our trained systems with existing methods such as COMET, BERT-Scorer, and LABSE, and found that the LLM-based evaluator (LLaMA-2-13B) achieves a comparable or higher overall correlation with human judgments for the considered Indian language pairs.

Neural abstractive summarization models are prone to generate content inconsistent with the source document, i.e. unfaithful. Existing automatic metrics do not capture such mistakes effectively. We tackle the problem of evaluating faithfulness of a generated summary given its source document. We first collected human annotations of faithfulness for outputs from numerous models on two datasets. We find that current models exhibit a trade-off between abstractiveness and faithfulness: outputs with less word overlap with the source document are more likely to be unfaithful. Next, we propose an automatic question answering (QA) based metric for faithfulness, FEQA, which leverages recent advances in reading comprehension. Given question-answer pairs generated from the summary, a QA model extracts answers from the document; non-matched answers indicate unfaithful information in the summary. Among metrics based on word overlap, embedding similarity, and learned language understanding models, our QA-based metric has significantly higher correlation with human faithfulness scores, especially on highly abstractive summaries.

Recent multimodal image generators such as GPT-4o, Gemini 2.0 Flash, and Gemini 2.5 Pro excel at following complex instructions, editing images and maintaining concept consistency. However, they are still evaluated by disjoint toolkits: text-to-image (T2I) benchmarks that lacks multi-modal conditioning, and customized image generation benchmarks that overlook compositional semantics and common knowledge. We propose MMIG-Bench, a comprehensive Multi-Modal Image Generation Benchmark that unifies these tasks by pairing 4,850 richly annotated text prompts with 1,750 multi-view reference images across 380 subjects, spanning humans, animals, objects, and artistic styles. MMIG-Bench is equipped with a three-level evaluation framework: (1) low-level metrics for visual artifacts and identity preservation of objects; (2) novel Aspect Matching Score (AMS): a VQA-based mid-level metric that delivers fine-grained prompt-image alignment and shows strong correlation with human judgments; and (3) high-level metrics for aesthetics and human preference. Using MMIG-Bench, we benchmark 17 state-of-the-art models, including Gemini 2.5 Pro, FLUX, DreamBooth, and IP-Adapter, and validate our metrics with 32k human ratings, yielding in-depth insights into architecture and data design. We will release the dataset and evaluation code to foster rigorous, unified evaluation and accelerate future innovations in multi-modal image generation.

Given the widespread adoption and usage of Large Language Models (LLMs), it is crucial to have flexible and interpretable evaluations of their instruction-following ability. Preference judgments between model outputs have become the de facto evaluation standard, despite distilling complex, multi-faceted preferences into a single ranking. Furthermore, as human annotation is slow and costly, LLMs are increasingly used to make these judgments, at the expense of reliability and interpretability. In this work, we propose TICK (Targeted Instruct-evaluation with ChecKlists), a fully automated, interpretable evaluation protocol that structures evaluations with LLM-generated, instruction-specific checklists. We first show that, given an instruction, LLMs can reliably produce high-quality, tailored evaluation checklists that decompose the instruction into a series of YES/NO questions. Each question asks whether a candidate response meets a specific requirement of the instruction. We demonstrate that using TICK leads to a significant increase (46.4% to 52.2%) in the frequency of exact agreements between LLM judgements and human preferences, as compared to having an LLM directly score an output. We then show that STICK (Self-TICK) can be used to improve generation quality across multiple benchmarks via self-refinement and Best-of-N selection. STICK self-refinement on LiveBench reasoning tasks leads to an absolute gain of +7.8%, whilst Best-of-N selection with STICK attains +6.3% absolute improvement on the real-world instruction dataset, WildBench. In light of this, structured, multi-faceted self-improvement is shown to be a promising way to further advance LLM capabilities. Finally, by providing LLM-generated checklists to human evaluators tasked with directly scoring LLM responses to WildBench instructions, we notably increase inter-annotator agreement (0.194 to 0.256).

Aesthetics serve as an implicit and important criterion in song generation tasks that reflect human perception beyond objective metrics. However, evaluating the aesthetics of generated songs remains a fundamental challenge, as the appreciation of music is highly subjective. Existing evaluation metrics, such as embedding-based distances, are limited in reflecting the subjective and perceptual aspects that define musical appeal. To address this issue, we introduce SongEval, the first open-source, large-scale benchmark dataset for evaluating the aesthetics of full-length songs. SongEval includes over 2,399 songs in full length, summing up to more than 140 hours, with aesthetic ratings from 16 professional annotators with musical backgrounds. Each song is evaluated across five key dimensions: overall coherence, memorability, naturalness of vocal breathing and phrasing, clarity of song structure, and overall musicality. The dataset covers both English and Chinese songs, spanning nine mainstream genres. Moreover, to assess the effectiveness of song aesthetic evaluation, we conduct experiments using SongEval to predict aesthetic scores and demonstrate better performance than existing objective evaluation metrics in predicting human-perceived musical quality.

Pre-trained large language models (LLMs) can be tailored to adhere to human instructions through instruction tuning. However, due to shifts in the distribution of test-time data, they may not always execute instructions accurately, potentially generating factual errors or misaligned content when acting as chat assistants. To enhance the reliability of LLMs in following instructions, we propose the study of selective instruction following, whereby the system declines to execute instructions if the anticipated response quality is low. We train judge models that can predict numerical quality scores for model responses. To address data scarcity, we introduce Self-J, a novel self-training framework for developing judge models without needing human-annotated quality scores. Our method leverages the model's inherent self-evaluation capability to extract information about response quality from labeled instruction-tuning data. It incorporates a gold reference answer to facilitate self-evaluation and recalibrates by assessing the semantic similarity between the response sample and the gold reference. During the training phase, we implement self-distillation as a regularization technique to enhance the capability of reference-free estimation. To validate alignment evaluation on general instruction-following tasks, we collect large-scale high-quality instructions from Hugging Face for model training and evaluation. Extensive experiments on five open-source models show that our method correlates much more with GPT-4 than strong baselines, e.g., supervised models distilled from GPT-4 and GPT-3.5-turbo. Our analysis shows our model's strong generalization across domains. Additionally, our judge models serve as good reward models, e.g., boosting WizardLM-13B-V1.2 from 89.17 to 92.48 and from 12.03 to 15.90 in version v1 and v2 of AlpacaEval respectively using best-of-32 sampling with our judge models.

Chatbots are designed to carry out human-like conversations across different domains, such as general chit-chat, knowledge exchange, and persona-grounded conversations. To measure the quality of such conversational agents, a dialogue evaluator is expected to conduct assessment across domains as well. However, most of the state-of-the-art automatic dialogue evaluation metrics (ADMs) are not designed for multi-domain evaluation. We are motivated to design a general and robust framework, MDD-Eval, to address the problem. Specifically, we first train a teacher evaluator with human-annotated data to acquire a rating skill to tell good dialogue responses from bad ones in a particular domain and then, adopt a self-training strategy to train a new evaluator with teacher-annotated multi-domain data, that helps the new evaluator to generalize across multiple domains. MDD-Eval is extensively assessed on six dialogue evaluation benchmarks. Empirical results show that the MDD-Eval framework achieves a strong performance with an absolute improvement of 7% over the state-of-the-art ADMs in terms of mean Spearman correlation scores across all the evaluation benchmarks.

Large Language Models (LLMs) have demonstrated impressive performance on Natural Language Processing (NLP) tasks, such as Question Answering, Summarization, and Classification. The use of LLMs as evaluators, that can rank or score the output of other models (usually LLMs) has become increasingly popular, due to the limitations of current evaluation techniques including the lack of appropriate benchmarks, metrics, cost, and access to human annotators. While LLMs are capable of handling approximately 100 languages, the majority of languages beyond the top 20 lack systematic evaluation across various tasks, metrics, and benchmarks. This creates an urgent need to scale up multilingual evaluation to ensure a precise understanding of LLM performance across diverse languages. LLM-based evaluators seem like the perfect solution to this problem, as they do not require human annotators, human-created references, or benchmarks and can theoretically be used to evaluate any language covered by the LLM. In this paper, we investigate whether LLM-based evaluators can help scale up multilingual evaluation. Specifically, we calibrate LLM-based evaluation against 20k human judgments of five metrics across three text-generation tasks in eight languages. Our findings indicate that LLM-based evaluators may exhibit bias towards higher scores and should be used with caution and should always be calibrated with a dataset of native speaker judgments, particularly in low-resource and non-Latin script languages.