venue stringclasses 1
value | title stringlengths 18 162 | abstract stringlengths 252 1.89k | doc_id stringlengths 32 32 | publication_year int64 2.02k 2.02k | sentences listlengths 1 13 | events listlengths 1 24 | document listlengths 50 348 |
|---|---|---|---|---|---|---|---|
ACL | Studying Summarization Evaluation Metrics in the Appropriate Scoring Range | In summarization, automatic evaluation metrics are usually compared based on their ability to correlate with human judgments. Unfortunately, the few existing human judgment datasets have been created as by-products of the manual evaluations performed during the DUC/TAC shared tasks. However, modern systems are typicall... | 487754b0963cb94bbf865cb42b7df0b0 | 2,019 | [
"in summarization , automatic evaluation metrics are usually compared based on their ability to correlate with human judgments .",
"unfortunately , the few existing human judgment datasets have been created as by - products of the manual evaluations performed during the duc / tac shared tasks .",
"however , mod... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "automatic evaluation metrics",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"automatic",
"evaluation",
"metrics"
],
"offsets": [
3,
4,
... | [
"in",
"summarization",
",",
"automatic",
"evaluation",
"metrics",
"are",
"usually",
"compared",
"based",
"on",
"their",
"ability",
"to",
"correlate",
"with",
"human",
"judgments",
".",
"unfortunately",
",",
"the",
"few",
"existing",
"human",
"judgment",
"datasets"... |
ACL | Temporal Common Sense Acquisition with Minimal Supervision | Temporal common sense (e.g., duration and frequency of events) is crucial for understanding natural language. However, its acquisition is challenging, partly because such information is often not expressed explicitly in text, and human annotation on such concepts is costly. This work proposes a novel sequence modeling ... | 2468d4b01a0bef23a6515909b3f7c29c | 2,020 | [
"temporal common sense ( e . g . , duration and frequency of events ) is crucial for understanding natural language .",
"however , its acquisition is challenging , partly because such information is often not expressed explicitly in text , and human annotation on such concepts is costly .",
"this work proposes ... | [
{
"event_type": "RWF",
"arguments": [
{
"text": "challenging",
"nugget_type": "WEA",
"argument_type": "Fault",
"tokens": [
"challenging"
],
"offsets": [
27
]
}
],
"trigger": {
"text": "challenging",
"... | [
"temporal",
"common",
"sense",
"(",
"e",
".",
"g",
".",
",",
"duration",
"and",
"frequency",
"of",
"events",
")",
"is",
"crucial",
"for",
"understanding",
"natural",
"language",
".",
"however",
",",
"its",
"acquisition",
"is",
"challenging",
",",
"partly",
... |
ACL | Modeling Intra-Relation in Math Word Problems with Different Functional Multi-Head Attentions | Several deep learning models have been proposed for solving math word problems (MWPs) automatically. Although these models have the ability to capture features without manual efforts, their approaches to capturing features are not specifically designed for MWPs. To utilize the merits of deep learning models with simult... | 296f04e4eb0e228328ba71f8f24b1638 | 2,019 | [
"several deep learning models have been proposed for solving math word problems ( mwps ) automatically .",
"although these models have the ability to capture features without manual efforts , their approaches to capturing features are not specifically designed for mwps .",
"to utilize the merits of deep learnin... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "deep learning models",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"deep",
"learning",
"models"
],
"offsets": [
1,
2,
3
... | [
"several",
"deep",
"learning",
"models",
"have",
"been",
"proposed",
"for",
"solving",
"math",
"word",
"problems",
"(",
"mwps",
")",
"automatically",
".",
"although",
"these",
"models",
"have",
"the",
"ability",
"to",
"capture",
"features",
"without",
"manual",
... |
ACL | Uncertainty Determines the Adequacy of the Mode and the Tractability of Decoding in Sequence-to-Sequence Models | In many natural language processing (NLP) tasks the same input (e.g. source sentence) can have multiple possible outputs (e.g. translations). To analyze how this ambiguity (also known as intrinsic uncertainty) shapes the distribution learned by neural sequence models we measure sentence-level uncertainty by computing t... | b41930d02e8edaa961fd5362dd3c08ba | 2,022 | [
"in many natural language processing ( nlp ) tasks the same input ( e . g . source sentence ) can have multiple possible outputs ( e . g . translations ) .",
"to analyze how this ambiguity ( also known as intrinsic uncertainty ) shapes the distribution learned by neural sequence models we measure sentence - level... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "natural language processing",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"natural",
"language",
"processing"
],
"offsets": [
2,
3,
... | [
"in",
"many",
"natural",
"language",
"processing",
"(",
"nlp",
")",
"tasks",
"the",
"same",
"input",
"(",
"e",
".",
"g",
".",
"source",
"sentence",
")",
"can",
"have",
"multiple",
"possible",
"outputs",
"(",
"e",
".",
"g",
".",
"translations",
")",
"."... |
ACL | Crossing Variational Autoencoders for Answer Retrieval | Answer retrieval is to find the most aligned answer from a large set of candidates given a question. Learning vector representations of questions/answers is the key factor. Question-answer alignment and question/answer semantics are two important signals for learning the representations. Existing methods learned semant... | 7168087ae03263276d7a0ae5ba741f75 | 2,020 | [
"answer retrieval is to find the most aligned answer from a large set of candidates given a question .",
"learning vector representations of questions / answers is the key factor .",
"question - answer alignment and question / answer semantics are two important signals for learning the representations .",
"ex... | [
{
"event_type": "RWF",
"arguments": [
{
"text": "too separate",
"nugget_type": "WEA",
"argument_type": "Fault",
"tokens": [
"too",
"separate"
],
"offsets": [
95,
96
]
},
{
"text": "align... | [
"answer",
"retrieval",
"is",
"to",
"find",
"the",
"most",
"aligned",
"answer",
"from",
"a",
"large",
"set",
"of",
"candidates",
"given",
"a",
"question",
".",
"learning",
"vector",
"representations",
"of",
"questions",
"/",
"answers",
"is",
"the",
"key",
"fa... |
ACL | Fantastic Questions and Where to Find Them: FairytaleQA – An Authentic Dataset for Narrative Comprehension | Question answering (QA) is a fundamental means to facilitate assessment and training of narrative comprehension skills for both machines and young children, yet there is scarcity of high-quality QA datasets carefully designed to serve this purpose. In particular, existing datasets rarely distinguish fine-grained readin... | ad500c6ff0fe9442a3aba99cfe4f5575 | 2,022 | [
"question answering ( qa ) is a fundamental means to facilitate assessment and training of narrative comprehension skills for both machines and young children , yet there is scarcity of high - quality qa datasets carefully designed to serve this purpose .",
"in particular , existing datasets rarely distinguish fi... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "question answering",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"question",
"answering"
],
"offsets": [
0,
1
]
}
],
"trigger"... | [
"question",
"answering",
"(",
"qa",
")",
"is",
"a",
"fundamental",
"means",
"to",
"facilitate",
"assessment",
"and",
"training",
"of",
"narrative",
"comprehension",
"skills",
"for",
"both",
"machines",
"and",
"young",
"children",
",",
"yet",
"there",
"is",
"sc... |
ACL | Image-Chat: Engaging Grounded Conversations | To achieve the long-term goal of machines being able to engage humans in conversation, our models should captivate the interest of their speaking partners. Communication grounded in images, whereby a dialogue is conducted based on a given photo, is a setup naturally appealing to humans (Hu et al., 2014). In this work w... | bf54a7de259a90356ed58550017e77e4 | 2,020 | [
"to achieve the long - term goal of machines being able to engage humans in conversation , our models should captivate the interest of their speaking partners .",
"communication grounded in images , whereby a dialogue is conducted based on a given photo , is a setup naturally appealing to humans ( hu et al . , 20... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "machines being able to engage humans in conversation",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"machines",
"being",
"able",
"to",
"engage",
... | [
"to",
"achieve",
"the",
"long",
"-",
"term",
"goal",
"of",
"machines",
"being",
"able",
"to",
"engage",
"humans",
"in",
"conversation",
",",
"our",
"models",
"should",
"captivate",
"the",
"interest",
"of",
"their",
"speaking",
"partners",
".",
"communication",... |
ACL | CBLUE: A Chinese Biomedical Language Understanding Evaluation Benchmark | Artificial Intelligence (AI), along with the recent progress in biomedical language understanding, is gradually offering great promise for medical practice. With the development of biomedical language understanding benchmarks, AI applications are widely used in the medical field. However, most benchmarks are limited to... | b13a4908f3f2001613ec52710a688146 | 2,022 | [
"artificial intelligence ( ai ) , along with the recent progress in biomedical language understanding , is gradually offering great promise for medical practice .",
"with the development of biomedical language understanding benchmarks , ai applications are widely used in the medical field .",
"however , most be... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "artificial intelligence",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"artificial",
"intelligence"
],
"offsets": [
0,
1
]
}
],
... | [
"artificial",
"intelligence",
"(",
"ai",
")",
",",
"along",
"with",
"the",
"recent",
"progress",
"in",
"biomedical",
"language",
"understanding",
",",
"is",
"gradually",
"offering",
"great",
"promise",
"for",
"medical",
"practice",
".",
"with",
"the",
"developme... |
ACL | Lightweight Cross-Lingual Sentence Representation Learning | Large-scale models for learning fixed-dimensional cross-lingual sentence representations like Large-scale models for learning fixed-dimensional cross-lingual sentence representations like LASER (Artetxe and Schwenk, 2019b) lead to significant improvement in performance on downstream tasks. However, further increases an... | ff2eff6801142da9f471d083c0310181 | 2,021 | [
"large - scale models for learning fixed - dimensional cross - lingual sentence representations like large - scale models for learning fixed - dimensional cross - lingual sentence representations like laser",
"( artetxe and schwenk , 2019b ) lead to significant improvement in performance on downstream tasks .",
... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "downstream tasks",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"downstream",
"tasks"
],
"offsets": [
45,
46
]
}
],
"trigger": ... | [
"large",
"-",
"scale",
"models",
"for",
"learning",
"fixed",
"-",
"dimensional",
"cross",
"-",
"lingual",
"sentence",
"representations",
"like",
"large",
"-",
"scale",
"models",
"for",
"learning",
"fixed",
"-",
"dimensional",
"cross",
"-",
"lingual",
"sentence",... |
ACL | Unified Speech-Text Pre-training for Speech Translation and Recognition | In this work, we describe a method to jointly pre-train speech and text in an encoder-decoder modeling framework for speech translation and recognition. The proposed method utilizes multi-task learning to integrate four self-supervised and supervised subtasks for cross modality learning. A self-supervised speech subtas... | 68beb39d47da3864f676700221819112 | 2,022 | [
"in this work , we describe a method to jointly pre - train speech and text in an encoder - decoder modeling framework for speech translation and recognition .",
"the proposed method utilizes multi - task learning to integrate four self - supervised and supervised subtasks for cross modality learning .",
"a sel... | [
{
"event_type": "MDS",
"arguments": [
{
"text": "multi - task learning",
"nugget_type": "APP",
"argument_type": "TriedComponent",
"tokens": [
"multi",
"-",
"task",
"learning"
],
"offsets": [
33,
... | [
"in",
"this",
"work",
",",
"we",
"describe",
"a",
"method",
"to",
"jointly",
"pre",
"-",
"train",
"speech",
"and",
"text",
"in",
"an",
"encoder",
"-",
"decoder",
"modeling",
"framework",
"for",
"speech",
"translation",
"and",
"recognition",
".",
"the",
"pr... |
ACL | Dialogue-Based Relation Extraction | We present the first human-annotated dialogue-based relation extraction (RE) dataset DialogRE, aiming to support the prediction of relation(s) between two arguments that appear in a dialogue. We further offer DialogRE as a platform for studying cross-sentence RE as most facts span multiple sentences. We argue that spea... | 8eb9ec741d73cf2fdc66bf4e00937e0f | 2,020 | [
"we present the first human - annotated dialogue - based relation extraction ( re ) dataset dialogre , aiming to support the prediction of relation ( s ) between two arguments that appear in a dialogue .",
"we further offer dialogre as a platform for studying cross - sentence re as most facts span multiple senten... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
0
]
},
{
"text": "first human - annotated dialogue - based relation e... | [
"we",
"present",
"the",
"first",
"human",
"-",
"annotated",
"dialogue",
"-",
"based",
"relation",
"extraction",
"(",
"re",
")",
"dataset",
"dialogre",
",",
"aiming",
"to",
"support",
"the",
"prediction",
"of",
"relation",
"(",
"s",
")",
"between",
"two",
"... |
ACL | MobileBERT: a Compact Task-Agnostic BERT for Resource-Limited Devices | Natural Language Processing (NLP) has recently achieved great success by using huge pre-trained models with hundreds of millions of parameters. However, these models suffer from heavy model sizes and high latency such that they cannot be deployed to resource-limited mobile devices. In this paper, we propose MobileBERT ... | 511ba6aecfe042b963b5e86370536611 | 2,020 | [
"natural language processing ( nlp ) has recently achieved great success by using huge pre - trained models with hundreds of millions of parameters .",
"however , these models suffer from heavy model sizes and high latency such that they cannot be deployed to resource - limited mobile devices .",
"in this paper... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "nlp",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"nlp"
],
"offsets": [
4
]
}
],
"trigger": {
"text": "has",
"tokens": [
"has"... | [
"natural",
"language",
"processing",
"(",
"nlp",
")",
"has",
"recently",
"achieved",
"great",
"success",
"by",
"using",
"huge",
"pre",
"-",
"trained",
"models",
"with",
"hundreds",
"of",
"millions",
"of",
"parameters",
".",
"however",
",",
"these",
"models",
... |
ACL | Feeding What You Need by Understanding What You Learned | Machine Reading Comprehension (MRC) reveals the ability to understand a given text passage and answer questions based on it. Existing research works in MRC rely heavily on large-size models and corpus to improve the performance evaluated by metrics such as Exact Match (EM) and F1. However, such a paradigm lacks suffici... | 19734dc50927f310d664aba28a178319 | 2,022 | [
"machine reading comprehension ( mrc ) reveals the ability to understand a given text passage and answer questions based on it .",
"existing research works in mrc rely heavily on large - size models and corpus to improve the performance evaluated by metrics such as exact match ( em ) and f1 .",
"however , such ... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "machine reading comprehension",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"machine",
"reading",
"comprehension"
],
"offsets": [
0,
1,
... | [
"machine",
"reading",
"comprehension",
"(",
"mrc",
")",
"reveals",
"the",
"ability",
"to",
"understand",
"a",
"given",
"text",
"passage",
"and",
"answer",
"questions",
"based",
"on",
"it",
".",
"existing",
"research",
"works",
"in",
"mrc",
"rely",
"heavily",
... |
ACL | Developmental Negation Processing in Transformer Language Models | Reasoning using negation is known to be difficult for transformer-based language models. While previous studies have used the tools of psycholinguistics to probe a transformer’s ability to reason over negation, none have focused on the types of negation studied in developmental psychology. We explore how well transform... | ef47600600700a317680db94f2b3252f | 2,022 | [
"reasoning using negation is known to be difficult for transformer - based language models .",
"while previous studies have used the tools of psycholinguistics to probe a transformer ’ s ability to reason over negation , none have focused on the types of negation studied in developmental psychology .",
"we expl... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "transformer - based language models",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"transformer",
"-",
"based",
"language",
"models"
],
"of... | [
"reasoning",
"using",
"negation",
"is",
"known",
"to",
"be",
"difficult",
"for",
"transformer",
"-",
"based",
"language",
"models",
".",
"while",
"previous",
"studies",
"have",
"used",
"the",
"tools",
"of",
"psycholinguistics",
"to",
"probe",
"a",
"transformer",... |
ACL | “The Boating Store Had Its Best Sail Ever”: Pronunciation-attentive Contextualized Pun Recognition | Humor plays an important role in human languages and it is essential to model humor when building intelligence systems. Among different forms of humor, puns perform wordplay for humorous effects by employing words with double entendre and high phonetic similarity. However, identifying and modeling puns are challenging ... | d404121e74488028f6a7b1980c5f0d03 | 2,020 | [
"humor plays an important role in human languages and it is essential to model humor when building intelligence systems .",
"among different forms of humor , puns perform wordplay for humorous effects by employing words with double entendre and high phonetic similarity .",
"however , identifying and modeling pu... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "humor",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"humor"
],
"offsets": [
0
]
}
],
"trigger": {
"text": "plays",
"tokens": [
... | [
"humor",
"plays",
"an",
"important",
"role",
"in",
"human",
"languages",
"and",
"it",
"is",
"essential",
"to",
"model",
"humor",
"when",
"building",
"intelligence",
"systems",
".",
"among",
"different",
"forms",
"of",
"humor",
",",
"puns",
"perform",
"wordplay... |
ACL | Beyond Sentence-Level End-to-End Speech Translation: Context Helps | Document-level contextual information has shown benefits to text-based machine translation, but whether and how context helps end-to-end (E2E) speech translation (ST) is still under-studied. We fill this gap through extensive experiments using a simple concatenation-based context-aware ST model, paired with adaptive fe... | b8e0f36c7739af9cd877c1379de41c8a | 2,021 | [
"document - level contextual information has shown benefits to text - based machine translation , but whether and how context helps end - to - end ( e2e ) speech translation ( st ) is still under - studied .",
"we fill this gap through extensive experiments using a simple concatenation - based context - aware st ... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "text - based machine translation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"text",
"-",
"based",
"machine",
"translation"
],
"offsets"... | [
"document",
"-",
"level",
"contextual",
"information",
"has",
"shown",
"benefits",
"to",
"text",
"-",
"based",
"machine",
"translation",
",",
"but",
"whether",
"and",
"how",
"context",
"helps",
"end",
"-",
"to",
"-",
"end",
"(",
"e2e",
")",
"speech",
"tran... |
ACL | Few-Shot Tabular Data Enrichment Using Fine-Tuned Transformer Architectures | The enrichment of tabular datasets using external sources has gained significant attention in recent years. Existing solutions, however, either ignore external unstructured data completely or devise dataset-specific solutions. In this study we proposed Few-Shot Transformer based Enrichment (FeSTE), a generic and robust... | fa4a78d6e03ad0effbc064e01bbce5fc | 2,022 | [
"the enrichment of tabular datasets using external sources has gained significant attention in recent years .",
"existing solutions , however , either ignore external unstructured data completely or devise dataset - specific solutions .",
"in this study we proposed few - shot transformer based enrichment ( fest... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "tabular datasets",
"nugget_type": "DST",
"argument_type": "Target",
"tokens": [
"tabular",
"datasets"
],
"offsets": [
3,
4
]
}
],
"trigger": {
... | [
"the",
"enrichment",
"of",
"tabular",
"datasets",
"using",
"external",
"sources",
"has",
"gained",
"significant",
"attention",
"in",
"recent",
"years",
".",
"existing",
"solutions",
",",
"however",
",",
"either",
"ignore",
"external",
"unstructured",
"data",
"comp... |
ACL | Symbolic Inductive Bias for Visually Grounded Learning of Spoken Language | A widespread approach to processing spoken language is to first automatically transcribe it into text. An alternative is to use an end-to-end approach: recent works have proposed to learn semantic embeddings of spoken language from images with spoken captions, without an intermediate transcription step. We propose to u... | 88069dd107a5732067679bb1e06a76c6 | 2,019 | [
"a widespread approach to processing spoken language is to first automatically transcribe it into text .",
"an alternative is to use an end - to - end approach : recent works have proposed to learn semantic embeddings of spoken language from images with spoken captions , without an intermediate transcription step... | [
{
"event_type": "WKS",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Researcher",
"tokens": [
"we"
],
"offsets": [
72
]
},
{
"text": "three - task architecture",
"nugget_type... | [
"a",
"widespread",
"approach",
"to",
"processing",
"spoken",
"language",
"is",
"to",
"first",
"automatically",
"transcribe",
"it",
"into",
"text",
".",
"an",
"alternative",
"is",
"to",
"use",
"an",
"end",
"-",
"to",
"-",
"end",
"approach",
":",
"recent",
"... |
ACL | Tchebycheff Procedure for Multi-task Text Classification | Multi-task Learning methods have achieved great progress in text classification. However, existing methods assume that multi-task text classification problems are convex multiobjective optimization problems, which is unrealistic in real-world applications. To address this issue, this paper presents a novel Tchebycheff ... | ae371754501712318cdfda6da51d1b64 | 2,020 | [
"multi - task learning methods have achieved great progress in text classification .",
"however , existing methods assume that multi - task text classification problems are convex multiobjective optimization problems , which is unrealistic in real - world applications .",
"to address this issue , this paper pre... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "multi - task learning methods",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"multi",
"-",
"task",
"learning",
"methods"
],
"offsets": [
... | [
"multi",
"-",
"task",
"learning",
"methods",
"have",
"achieved",
"great",
"progress",
"in",
"text",
"classification",
".",
"however",
",",
"existing",
"methods",
"assume",
"that",
"multi",
"-",
"task",
"text",
"classification",
"problems",
"are",
"convex",
"mult... |
ACL | Embracing Ambiguity: Shifting the Training Target of NLI Models | Natural Language Inference (NLI) datasets contain examples with highly ambiguous labels. While many research works do not pay much attention to this fact, several recent efforts have been made to acknowledge and embrace the existence of ambiguity, such as UNLI and ChaosNLI. In this paper, we explore the option of train... | 5ad3e1fe9612944430ccb90b4cbf9d69 | 2,021 | [
"natural language inference ( nli ) datasets contain examples with highly ambiguous labels .",
"while many research works do not pay much attention to this fact , several recent efforts have been made to acknowledge and embrace the existence of ambiguity , such as unli and chaosnli .",
"in this paper , we explo... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "natural language inference ( nli ) datasets",
"nugget_type": "DST",
"argument_type": "Target",
"tokens": [
"natural",
"language",
"inference",
"(",
"nli",
")",
... | [
"natural",
"language",
"inference",
"(",
"nli",
")",
"datasets",
"contain",
"examples",
"with",
"highly",
"ambiguous",
"labels",
".",
"while",
"many",
"research",
"works",
"do",
"not",
"pay",
"much",
"attention",
"to",
"this",
"fact",
",",
"several",
"recent",... |
ACL | Turn the Combination Lock: Learnable Textual Backdoor Attacks via Word Substitution | Recent studies show that neural natural language processing (NLP) models are vulnerable to backdoor attacks. Injected with backdoors, models perform normally on benign examples but produce attacker-specified predictions when the backdoor is activated, presenting serious security threats to real-world applications. Sinc... | ad85d1bf129dfe363cb7280f7b02c171 | 2,021 | [
"recent studies show that neural natural language processing ( nlp ) models are vulnerable to backdoor attacks .",
"injected with backdoors , models perform normally on benign examples but produce attacker - specified predictions when the backdoor is activated , presenting serious security threats to real - world... | [
{
"event_type": "RWF",
"arguments": [
{
"text": "neural natural language processing models",
"nugget_type": "APP",
"argument_type": "Concern",
"tokens": [
"neural",
"natural",
"language",
"processing",
"models"
]... | [
"recent",
"studies",
"show",
"that",
"neural",
"natural",
"language",
"processing",
"(",
"nlp",
")",
"models",
"are",
"vulnerable",
"to",
"backdoor",
"attacks",
".",
"injected",
"with",
"backdoors",
",",
"models",
"perform",
"normally",
"on",
"benign",
"examples... |
ACL | Latent Variable Model for Multi-modal Translation | In this work, we propose to model the interaction between visual and textual features for multi-modal neural machine translation (MMT) through a latent variable model. This latent variable can be seen as a multi-modal stochastic embedding of an image and its description in a foreign language. It is used in a target-lan... | 67cf55b9e9bc5924d9631c444d74122f | 2,019 | [
"in this work , we propose to model the interaction between visual and textual features for multi - modal neural machine translation ( mmt ) through a latent variable model .",
"this latent variable can be seen as a multi - modal stochastic embedding of an image and its description in a foreign language .",
"it... | [
{
"event_type": "WKS",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Researcher",
"tokens": [
"we"
],
"offsets": [
4
]
},
{
"text": "multi - modal neural machine translation",
... | [
"in",
"this",
"work",
",",
"we",
"propose",
"to",
"model",
"the",
"interaction",
"between",
"visual",
"and",
"textual",
"features",
"for",
"multi",
"-",
"modal",
"neural",
"machine",
"translation",
"(",
"mmt",
")",
"through",
"a",
"latent",
"variable",
"mode... |
ACL | Towards Explainable NLP: A Generative Explanation Framework for Text Classification | Building explainable systems is a critical problem in the field of Natural Language Processing (NLP), since most machine learning models provide no explanations for the predictions. Existing approaches for explainable machine learning systems tend to focus on interpreting the outputs or the connections between inputs a... | b8096d9c6403467d207011915bb2306e | 2,019 | [
"building explainable systems is a critical problem in the field of natural language processing ( nlp ) , since most machine learning models provide no explanations for the predictions .",
"existing approaches for explainable machine learning systems tend to focus on interpreting the outputs or the connections be... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "natural language processing",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"natural",
"language",
"processing"
],
"offsets": [
11,
12,
... | [
"building",
"explainable",
"systems",
"is",
"a",
"critical",
"problem",
"in",
"the",
"field",
"of",
"natural",
"language",
"processing",
"(",
"nlp",
")",
",",
"since",
"most",
"machine",
"learning",
"models",
"provide",
"no",
"explanations",
"for",
"the",
"pre... |
ACL | Increasing Faithfulness in Knowledge-Grounded Dialogue with Controllable Features | Knowledge-grounded dialogue systems are intended to convey information that is based on evidence provided in a given source text. We discuss the challenges of training a generative neural dialogue model for such systems that is controlled to stay faithful to the evidence. Existing datasets contain a mix of conversation... | 0891d3012c0815d02eebc7c7be15b1ac | 2,021 | [
"knowledge - grounded dialogue systems are intended to convey information that is based on evidence provided in a given source text .",
"we discuss the challenges of training a generative neural dialogue model for such systems that is controlled to stay faithful to the evidence .",
"existing datasets contain a ... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "knowledge - grounded dialogue systems",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"knowledge",
"-",
"grounded",
"dialogue",
"systems"
],
... | [
"knowledge",
"-",
"grounded",
"dialogue",
"systems",
"are",
"intended",
"to",
"convey",
"information",
"that",
"is",
"based",
"on",
"evidence",
"provided",
"in",
"a",
"given",
"source",
"text",
".",
"we",
"discuss",
"the",
"challenges",
"of",
"training",
"a",
... |
ACL | Graph Neural Networks with Generated Parameters for Relation Extraction | In this paper, we propose a novel graph neural network with generated parameters (GP-GNNs). The parameters in the propagation module, i.e. the transition matrices used in message passing procedure, are produced by a generator taking natural language sentences as inputs. We verify GP-GNNs in relation extraction from tex... | 093cdd1123376f915d16b48d0fd41e0d | 2,019 | [
"in this paper , we propose a novel graph neural network with generated parameters ( gp - gnns ) .",
"the parameters in the propagation module , i . e . the transition matrices used in message passing procedure , are produced by a generator taking natural language sentences as inputs .",
"we verify gp - gnns in... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
4
]
},
{
"text": "graph neural network",
"nugget_type": "APP"... | [
"in",
"this",
"paper",
",",
"we",
"propose",
"a",
"novel",
"graph",
"neural",
"network",
"with",
"generated",
"parameters",
"(",
"gp",
"-",
"gnns",
")",
".",
"the",
"parameters",
"in",
"the",
"propagation",
"module",
",",
"i",
".",
"e",
".",
"the",
"tr... |
ACL | A Semantics-aware Transformer Model of Relation Linking for Knowledge Base Question Answering | Relation linking is a crucial component of Knowledge Base Question Answering systems. Existing systems use a wide variety of heuristics, or ensembles of multiple systems, heavily relying on the surface question text. However, the explicit semantic parse of the question is a rich source of relation information that is n... | d7957d4dc8cf07c10054ff40965f239d | 2,021 | [
"relation linking is a crucial component of knowledge base question answering systems .",
"existing systems use a wide variety of heuristics , or ensembles of multiple systems , heavily relying on the surface question text .",
"however , the explicit semantic parse of the question is a rich source of relation i... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "relation linking",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"relation",
"linking"
],
"offsets": [
0,
1
]
}
],
"trigger": {
... | [
"relation",
"linking",
"is",
"a",
"crucial",
"component",
"of",
"knowledge",
"base",
"question",
"answering",
"systems",
".",
"existing",
"systems",
"use",
"a",
"wide",
"variety",
"of",
"heuristics",
",",
"or",
"ensembles",
"of",
"multiple",
"systems",
",",
"h... |
ACL | ASPECTNEWS: Aspect-Oriented Summarization of News Documents | Generic summaries try to cover an entire document and query-based summaries try to answer document-specific questions. But real users’ needs often fall in between these extremes and correspond to aspects, high-level topics discussed among similar types of documents. In this paper, we collect a dataset of realistic aspe... | be932bceb0e2e8e3c11cb20992f049a7 | 2,022 | [
"generic summaries try to cover an entire document and query - based summaries try to answer document - specific questions .",
"but real users ’ needs often fall in between these extremes and correspond to aspects , high - level topics discussed among similar types of documents .",
"in this paper , we collect a... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
52
]
},
{
"text": "dataset of realistic aspect - oriented summaries",... | [
"generic",
"summaries",
"try",
"to",
"cover",
"an",
"entire",
"document",
"and",
"query",
"-",
"based",
"summaries",
"try",
"to",
"answer",
"document",
"-",
"specific",
"questions",
".",
"but",
"real",
"users",
"’",
"needs",
"often",
"fall",
"in",
"between",... |
ACL | Transkimmer: Transformer Learns to Layer-wise Skim | Transformer architecture has become the de-facto model for many machine learning tasks from natural language processing and computer vision. As such, improving its computational efficiency becomes paramount. One of the major computational inefficiency of Transformer based models is that they spend the identical amount ... | 0b340d4db5d460bf82ca71c69989be9f | 2,022 | [
"transformer architecture has become the de - facto model for many machine learning tasks from natural language processing and computer vision .",
"as such , improving its computational efficiency becomes paramount .",
"one of the major computational inefficiency of transformer based models is that they spend t... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "machine learning tasks",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"machine",
"learning",
"tasks"
],
"offsets": [
11,
12,
13
... | [
"transformer",
"architecture",
"has",
"become",
"the",
"de",
"-",
"facto",
"model",
"for",
"many",
"machine",
"learning",
"tasks",
"from",
"natural",
"language",
"processing",
"and",
"computer",
"vision",
".",
"as",
"such",
",",
"improving",
"its",
"computationa... |
ACL | Polyjuice: Generating Counterfactuals for Explaining, Evaluating, and Improving Models | While counterfactual examples are useful for analysis and training of NLP models, current generation methods either rely on manual labor to create very few counterfactuals, or only instantiate limited types of perturbations such as paraphrases or word substitutions. We present Polyjuice, a general-purpose counterfactua... | 4ec6280aa5b83537d731f4d923772ac1 | 2,021 | [
"while counterfactual examples are useful for analysis and training of nlp models , current generation methods either rely on manual labor to create very few counterfactuals , or only instantiate limited types of perturbations such as paraphrases or word substitutions .",
"we present polyjuice , a general - purpo... | [
{
"event_type": "RWF",
"arguments": [
{
"text": "very few counterfactuals",
"nugget_type": "WEA",
"argument_type": "Fault",
"tokens": [
"very",
"few",
"counterfactuals"
],
"offsets": [
23,
24,
2... | [
"while",
"counterfactual",
"examples",
"are",
"useful",
"for",
"analysis",
"and",
"training",
"of",
"nlp",
"models",
",",
"current",
"generation",
"methods",
"either",
"rely",
"on",
"manual",
"labor",
"to",
"create",
"very",
"few",
"counterfactuals",
",",
"or",
... |
ACL | From Simultaneous to Streaming Machine Translation by Leveraging Streaming History | Simultaneous machine translation has recently gained traction thanks to significant quality improvements and the advent of streaming applications. Simultaneous translation systems need to find a trade-off between translation quality and response time, and with this purpose multiple latency measures have been proposed. ... | 6547fcdb17989df75ff956ee23415010 | 2,022 | [
"simultaneous machine translation has recently gained traction thanks to significant quality improvements and the advent of streaming applications .",
"simultaneous translation systems need to find a trade - off between translation quality and response time , and with this purpose multiple latency measures have b... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "simultaneous machine translation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"simultaneous",
"machine",
"translation"
],
"offsets": [
0,
... | [
"simultaneous",
"machine",
"translation",
"has",
"recently",
"gained",
"traction",
"thanks",
"to",
"significant",
"quality",
"improvements",
"and",
"the",
"advent",
"of",
"streaming",
"applications",
".",
"simultaneous",
"translation",
"systems",
"need",
"to",
"find",... |
ACL | Norm-Based Curriculum Learning for Neural Machine Translation | A neural machine translation (NMT) system is expensive to train, especially with high-resource settings. As the NMT architectures become deeper and wider, this issue gets worse and worse. In this paper, we aim to improve the efficiency of training an NMT by introducing a novel norm-based curriculum learning method. We ... | ce4a670f516eef0e7b3b9e45b979f229 | 2,020 | [
"a neural machine translation ( nmt ) system is expensive to train , especially with high - resource settings .",
"as the nmt architectures become deeper and wider , this issue gets worse and worse .",
"in this paper , we aim to improve the efficiency of training an nmt by introducing a novel norm - based curri... | [
{
"event_type": "RWF",
"arguments": [
{
"text": "neural machine translation system",
"nugget_type": "APP",
"argument_type": "Concern",
"tokens": [
"neural",
"machine",
"translation",
"system"
],
"offsets": [
... | [
"a",
"neural",
"machine",
"translation",
"(",
"nmt",
")",
"system",
"is",
"expensive",
"to",
"train",
",",
"especially",
"with",
"high",
"-",
"resource",
"settings",
".",
"as",
"the",
"nmt",
"architectures",
"become",
"deeper",
"and",
"wider",
",",
"this",
... |
ACL | Guiding the Growth: Difficulty-Controllable Question Generation through Step-by-Step Rewriting | This paper explores the task of Difficulty-Controllable Question Generation (DCQG), which aims at generating questions with required difficulty levels. Previous research on this task mainly defines the difficulty of a question as whether it can be correctly answered by a Question Answering (QA) system, lacking interpre... | 9cebb4488ba399e66fdeb7351c4fc090 | 2,021 | [
"this paper explores the task of difficulty - controllable question generation ( dcqg ) , which aims at generating questions with required difficulty levels .",
"previous research on this task mainly defines the difficulty of a question as whether it can be correctly answered by a question answering ( qa ) system... | [
{
"event_type": "RWF",
"arguments": [
{
"text": "previous research",
"nugget_type": "APP",
"argument_type": "Concern",
"tokens": [
"previous",
"research"
],
"offsets": [
25,
26
]
},
{
"t... | [
"this",
"paper",
"explores",
"the",
"task",
"of",
"difficulty",
"-",
"controllable",
"question",
"generation",
"(",
"dcqg",
")",
",",
"which",
"aims",
"at",
"generating",
"questions",
"with",
"required",
"difficulty",
"levels",
".",
"previous",
"research",
"on",... |
ACL | Understanding Undesirable Word Embedding Associations | Word embeddings are often criticized for capturing undesirable word associations such as gender stereotypes. However, methods for measuring and removing such biases remain poorly understood. We show that for any embedding model that implicitly does matrix factorization, debiasing vectors post hoc using subspace project... | 649e8ff34ad67f491b65b18055866a17 | 2,019 | [
"word embeddings are often criticized for capturing undesirable word associations such as gender stereotypes .",
"however , methods for measuring and removing such biases remain poorly understood .",
"we show that for any embedding model that implicitly does matrix factorization , debiasing vectors post hoc usi... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "word embeddings",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"word",
"embeddings"
],
"offsets": [
0,
1
]
}
],
"trigger": {
... | [
"word",
"embeddings",
"are",
"often",
"criticized",
"for",
"capturing",
"undesirable",
"word",
"associations",
"such",
"as",
"gender",
"stereotypes",
".",
"however",
",",
"methods",
"for",
"measuring",
"and",
"removing",
"such",
"biases",
"remain",
"poorly",
"unde... |
ACL | Toward Comprehensive Understanding of a Sentiment Based on Human Motives | In sentiment detection, the natural language processing community has focused on determining holders, facets, and valences, but has paid little attention to the reasons for sentiment decisions. Our work considers human motives as the driver for human sentiments and addresses the problem of motive detection as the first... | becb993ed64f2b327191beeb9ccbf4a5 | 2,019 | [
"in sentiment detection , the natural language processing community has focused on determining holders , facets , and valences , but has paid little attention to the reasons for sentiment decisions .",
"our work considers human motives as the driver for human sentiments and addresses the problem of motive detecti... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "sentiment detection",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"sentiment",
"detection"
],
"offsets": [
1,
2
]
}
],
"trigge... | [
"in",
"sentiment",
"detection",
",",
"the",
"natural",
"language",
"processing",
"community",
"has",
"focused",
"on",
"determining",
"holders",
",",
"facets",
",",
"and",
"valences",
",",
"but",
"has",
"paid",
"little",
"attention",
"to",
"the",
"reasons",
"fo... |
ACL | A negative case analysis of visual grounding methods for VQA | Existing Visual Question Answering (VQA) methods tend to exploit dataset biases and spurious statistical correlations, instead of producing right answers for the right reasons. To address this issue, recent bias mitigation methods for VQA propose to incorporate visual cues (e.g., human attention maps) to better ground ... | 22f686944c025db2e2ea184cb22b777f | 2,020 | [
"existing visual question answering ( vqa ) methods tend to exploit dataset biases and spurious statistical correlations , instead of producing right answers for the right reasons .",
"to address this issue , recent bias mitigation methods for vqa propose to incorporate visual cues ( e . g . , human attention map... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "visual question answering methods",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"visual",
"question",
"answering",
"methods"
],
"offsets": [
... | [
"existing",
"visual",
"question",
"answering",
"(",
"vqa",
")",
"methods",
"tend",
"to",
"exploit",
"dataset",
"biases",
"and",
"spurious",
"statistical",
"correlations",
",",
"instead",
"of",
"producing",
"right",
"answers",
"for",
"the",
"right",
"reasons",
".... |
ACL | Sparse Progressive Distillation: Resolving Overfitting under Pretrain-and-Finetune Paradigm | Conventional wisdom in pruning Transformer-based language models is that pruning reduces the model expressiveness and thus is more likely to underfit rather than overfit. However, under the trending pretrain-and-finetune paradigm, we postulate a counter-traditional hypothesis, that is: pruning increases the risk of ove... | 73c696833523e880c1787882c8276481 | 2,022 | [
"conventional wisdom in pruning transformer - based language models is that pruning reduces the model expressiveness and thus is more likely to underfit rather than overfit .",
"however , under the trending pretrain - and - finetune paradigm , we postulate a counter - traditional hypothesis , that is : pruning in... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "transformer - based language models",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"transformer",
"-",
"based",
"language",
"models"
],
"of... | [
"conventional",
"wisdom",
"in",
"pruning",
"transformer",
"-",
"based",
"language",
"models",
"is",
"that",
"pruning",
"reduces",
"the",
"model",
"expressiveness",
"and",
"thus",
"is",
"more",
"likely",
"to",
"underfit",
"rather",
"than",
"overfit",
".",
"howeve... |
ACL | Target Conditioned Sampling: Optimizing Data Selection for Multilingual Neural Machine Translation | To improve low-resource Neural Machine Translation (NMT) with multilingual corpus, training on the most related high-resource language only is generally more effective than us- ing all data available (Neubig and Hu, 2018). However, it remains a question whether a smart data selection strategy can further improve low-re... | 9347bddd69fa66e65ef23b8f1109f6cc | 2,019 | [
"to improve low - resource neural machine translation ( nmt ) with multilingual corpus , training on the most related high - resource language only is generally more effective than us - ing all data available ( neubig and hu , 2018 ) .",
"however , it remains a question whether a smart data selection strategy can... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "low - resource neural machine translation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"low",
"-",
"resource",
"neural",
"machine",
"translatio... | [
"to",
"improve",
"low",
"-",
"resource",
"neural",
"machine",
"translation",
"(",
"nmt",
")",
"with",
"multilingual",
"corpus",
",",
"training",
"on",
"the",
"most",
"related",
"high",
"-",
"resource",
"language",
"only",
"is",
"generally",
"more",
"effective"... |
ACL | CLIP Models are Few-Shot Learners: Empirical Studies on VQA and Visual Entailment | CLIP has shown a remarkable zero-shot capability on a wide range of vision tasks. Previously, CLIP is only regarded as a powerful visual encoder. However, after being pre-trained by language supervision from a large amount of image-caption pairs, CLIP itself should also have acquired some few-shot abilities for vision-... | 4c0267d2b7bf2c37b87d6a20b909d00a | 2,022 | [
"clip has shown a remarkable zero - shot capability on a wide range of vision tasks .",
"previously , clip is only regarded as a powerful visual encoder .",
"however , after being pre - trained by language supervision from a large amount of image - caption pairs , clip itself should also have acquired some few ... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "clip",
"nugget_type": "MOD",
"argument_type": "Target",
"tokens": [
"clip"
],
"offsets": [
0
]
}
],
"trigger": {
"text": "shown",
"tokens": [
"... | [
"clip",
"has",
"shown",
"a",
"remarkable",
"zero",
"-",
"shot",
"capability",
"on",
"a",
"wide",
"range",
"of",
"vision",
"tasks",
".",
"previously",
",",
"clip",
"is",
"only",
"regarded",
"as",
"a",
"powerful",
"visual",
"encoder",
".",
"however",
",",
... |
ACL | From the Detection of Toxic Spans in Online Discussions to the Analysis of Toxic-to-Civil Transfer | We study the task of toxic spans detection, which concerns the detection of the spans that make a text toxic, when detecting such spans is possible. We introduce a dataset for this task, ToxicSpans, which we release publicly. By experimenting with several methods, we show that sequence labeling models perform best, but... | b271ee2489b1c38d73445defc430fc97 | 2,022 | [
"we study the task of toxic spans detection , which concerns the detection of the spans that make a text toxic , when detecting such spans is possible .",
"we introduce a dataset for this task , toxicspans , which we release publicly .",
"by experimenting with several methods , we show that sequence labeling mo... | [
{
"event_type": "WKS",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Researcher",
"tokens": [
"we"
],
"offsets": [
0
]
},
{
"text": "toxic spans detection",
"nugget_type": "T... | [
"we",
"study",
"the",
"task",
"of",
"toxic",
"spans",
"detection",
",",
"which",
"concerns",
"the",
"detection",
"of",
"the",
"spans",
"that",
"make",
"a",
"text",
"toxic",
",",
"when",
"detecting",
"such",
"spans",
"is",
"possible",
".",
"we",
"introduce"... |
ACL | A Semantic-based Method for Unsupervised Commonsense Question Answering | Unsupervised commonsense question answering is appealing since it does not rely on any labeled task data. Among existing work, a popular solution is to use pre-trained language models to score candidate choices directly conditioned on the question or context. However, such scores from language models can be easily affe... | 6291db754630fe8cc59435ea58c69883 | 2,021 | [
"unsupervised commonsense question answering is appealing since it does not rely on any labeled task data .",
"among existing work , a popular solution is to use pre - trained language models to score candidate choices directly conditioned on the question or context .",
"however , such scores from language mode... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "unsupervised commonsense question answering",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"unsupervised",
"commonsense",
"question",
"answering"
],
... | [
"unsupervised",
"commonsense",
"question",
"answering",
"is",
"appealing",
"since",
"it",
"does",
"not",
"rely",
"on",
"any",
"labeled",
"task",
"data",
".",
"among",
"existing",
"work",
",",
"a",
"popular",
"solution",
"is",
"to",
"use",
"pre",
"-",
"traine... |
ACL | GPT-too: A Language-Model-First Approach for AMR-to-Text Generation | Abstract Meaning Representations (AMRs) are broad-coverage sentence-level semantic graphs. Existing approaches to generating text from AMR have focused on training sequence-to-sequence or graph-to-sequence models on AMR annotated data only. In this paper, we propose an alternative approach that combines a strong pre-tr... | de0fcb7a0d9b9c23138d86db17136d89 | 2,020 | [
"abstract meaning representations ( amrs ) are broad - coverage sentence - level semantic graphs .",
"existing approaches to generating text from amr have focused on training sequence - to - sequence or graph - to - sequence models on amr annotated data only .",
"in this paper , we propose an alternative approa... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "abstract meaning representations",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"abstract",
"meaning",
"representations"
],
"offsets": [
0,
... | [
"abstract",
"meaning",
"representations",
"(",
"amrs",
")",
"are",
"broad",
"-",
"coverage",
"sentence",
"-",
"level",
"semantic",
"graphs",
".",
"existing",
"approaches",
"to",
"generating",
"text",
"from",
"amr",
"have",
"focused",
"on",
"training",
"sequence"... |
ACL | Mix and Match: Learning-free Controllable Text Generationusing Energy Language Models | Recent work on controlled text generation has either required attribute-based fine-tuning of the base language model (LM), or has restricted the parameterization of the attribute discriminator to be compatible with the base autoregressive LM. In this work, we propose Mix and Match LM, a global score-based alternative f... | 03d9477d848de642a07943b53f2ae11a | 2,022 | [
"recent work on controlled text generation has either required attribute - based fine - tuning of the base language model ( lm ) , or has restricted the parameterization of the attribute discriminator to be compatible with the base autoregressive lm .",
"in this work , we propose mix and match lm , a global score... | [
{
"event_type": "RWS",
"arguments": [
{
"text": "controlled text generation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"controlled",
"text",
"generation"
],
"offsets": [
3,
4,
... | [
"recent",
"work",
"on",
"controlled",
"text",
"generation",
"has",
"either",
"required",
"attribute",
"-",
"based",
"fine",
"-",
"tuning",
"of",
"the",
"base",
"language",
"model",
"(",
"lm",
")",
",",
"or",
"has",
"restricted",
"the",
"parameterization",
"o... |
ACL | StableMoE: Stable Routing Strategy for Mixture of Experts | The Mixture-of-Experts (MoE) technique can scale up the model size of Transformers with an affordable computational overhead. We point out that existing learning-to-route MoE methods suffer from the routing fluctuation issue, i.e., the target expert of the same input may change along with training, but only one expert ... | c3aae49ba675d4bb1410aa4734e9265f | 2,022 | [
"the mixture - of - experts ( moe ) technique can scale up the model size of transformers with an affordable computational overhead .",
"we point out that existing learning - to - route moe methods suffer from the routing fluctuation issue , i . e . , the target expert of the same input may change along with trai... | [
{
"event_type": "RWS",
"arguments": [
{
"text": "mixture - of - experts ( moe ) technique",
"nugget_type": "APP",
"argument_type": "Subject",
"tokens": [
"mixture",
"-",
"of",
"-",
"experts",
"(",
"mo... | [
"the",
"mixture",
"-",
"of",
"-",
"experts",
"(",
"moe",
")",
"technique",
"can",
"scale",
"up",
"the",
"model",
"size",
"of",
"transformers",
"with",
"an",
"affordable",
"computational",
"overhead",
".",
"we",
"point",
"out",
"that",
"existing",
"learning",... |
ACL | How Distributed are Distributed Representations? An Observation on the Locality of Syntactic Information in Verb Agreement Tasks | This work addresses the question of the localization of syntactic information encoded in the transformers representations. We tackle this question from two perspectives, considering the object-past participle agreement in French, by identifying, first, in which part of the sentence and, second, in which part of the rep... | c190c1f93bebba56582c748f50883344 | 2,022 | [
"this work addresses the question of the localization of syntactic information encoded in the transformers representations .",
"we tackle this question from two perspectives , considering the object - past participle agreement in french , by identifying , first , in which part of the sentence and , second , in wh... | [
{
"event_type": "WKS",
"arguments": [
{
"text": "question of the localization of syntactic information",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"question",
"of",
"the",
"localization",
"of",
... | [
"this",
"work",
"addresses",
"the",
"question",
"of",
"the",
"localization",
"of",
"syntactic",
"information",
"encoded",
"in",
"the",
"transformers",
"representations",
".",
"we",
"tackle",
"this",
"question",
"from",
"two",
"perspectives",
",",
"considering",
"t... |
ACL | Ruddit: Norms of Offensiveness for English Reddit Comments | On social media platforms, hateful and offensive language negatively impact the mental well-being of users and the participation of people from diverse backgrounds. Automatic methods to detect offensive language have largely relied on datasets with categorical labels. However, comments can vary in their degree of offen... | 956c6b8b8effff8592f2df9e400f2b96 | 2,021 | [
"on social media platforms , hateful and offensive language negatively impact the mental well - being of users and the participation of people from diverse backgrounds .",
"automatic methods to detect offensive language have largely relied on datasets with categorical labels .",
"however , comments can vary in ... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
53
]
},
{
"text": "first dataset",
"nugget_type": "DST",
... | [
"on",
"social",
"media",
"platforms",
",",
"hateful",
"and",
"offensive",
"language",
"negatively",
"impact",
"the",
"mental",
"well",
"-",
"being",
"of",
"users",
"and",
"the",
"participation",
"of",
"people",
"from",
"diverse",
"backgrounds",
".",
"automatic",... |
ACL | Coach: A Coarse-to-Fine Approach for Cross-domain Slot Filling | As an essential task in task-oriented dialog systems, slot filling requires extensive training data in a certain domain. However, such data are not always available. Hence, cross-domain slot filling has naturally arisen to cope with this data scarcity problem. In this paper, we propose a Coarse-to-fine approach (Coach)... | 3ed48f40902a00b298e050542220d8ee | 2,020 | [
"as an essential task in task - oriented dialog systems , slot filling requires extensive training data in a certain domain .",
"however , such data are not always available .",
"hence , cross - domain slot filling has naturally arisen to cope with this data scarcity problem .",
"in this paper , we propose a ... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "slot filling",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"slot",
"filling"
],
"offsets": [
11,
12
]
}
],
"trigger": {
... | [
"as",
"an",
"essential",
"task",
"in",
"task",
"-",
"oriented",
"dialog",
"systems",
",",
"slot",
"filling",
"requires",
"extensive",
"training",
"data",
"in",
"a",
"certain",
"domain",
".",
"however",
",",
"such",
"data",
"are",
"not",
"always",
"available"... |
ACL | Large Scale Multi-Actor Generative Dialog Modeling | Non-goal oriented dialog agents (i.e. chatbots) aim to produce varying and engaging conversations with a user; however, they typically exhibit either inconsistent personality across conversations or the average personality of all users. This paper addresses these issues by controlling an agent’s persona upon generation... | 54a8d5254b8db76aa072324ab4db438a | 2,020 | [
"non - goal oriented dialog agents ( i . e . chatbots ) aim to produce varying and engaging conversations with a user ; however , they typically exhibit either inconsistent personality across conversations or the average personality of all users .",
"this paper addresses these issues by controlling an agent ’ s p... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "non - goal oriented dialog agents",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"non",
"-",
"goal",
"oriented",
"dialog",
"agents"
],
... | [
"non",
"-",
"goal",
"oriented",
"dialog",
"agents",
"(",
"i",
".",
"e",
".",
"chatbots",
")",
"aim",
"to",
"produce",
"varying",
"and",
"engaging",
"conversations",
"with",
"a",
"user",
";",
"however",
",",
"they",
"typically",
"exhibit",
"either",
"incons... |
ACL | Local Languages, Third Spaces, and other High-Resource Scenarios | How can language technology address the diverse situations of the world’s languages? In one view, languages exist on a resource continuum and the challenge is to scale existing solutions, bringing under-resourced languages into the high-resource world. In another view, presented here, the world’s language ecology inclu... | d3bac761bfc244f710333105cb3f7ffc | 2,022 | [
"how can language technology address the diverse situations of the world ’ s languages ?",
"in one view , languages exist on a resource continuum and the challenge is to scale existing solutions , bringing under - resourced languages into the high - resource world .",
"in another view , presented here , the wor... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "diverse situations of the world ’ s languages",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"diverse",
"situations",
"of",
"the",
"world",
"’",
... | [
"how",
"can",
"language",
"technology",
"address",
"the",
"diverse",
"situations",
"of",
"the",
"world",
"’",
"s",
"languages",
"?",
"in",
"one",
"view",
",",
"languages",
"exist",
"on",
"a",
"resource",
"continuum",
"and",
"the",
"challenge",
"is",
"to",
... |
ACL | Multilingual Agreement for Multilingual Neural Machine Translation | Although multilingual neural machine translation (MNMT) enables multiple language translations, the training process is based on independent multilingual objectives. Most multilingual models can not explicitly exploit different language pairs to assist each other, ignoring the relationships among them. In this work, we... | d75d8fb666416dc15a541233ff6260dd | 2,021 | [
"although multilingual neural machine translation ( mnmt ) enables multiple language translations , the training process is based on independent multilingual objectives .",
"most multilingual models can not explicitly exploit different language pairs to assist each other , ignoring the relationships among them ."... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "multilingual neural machine translation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"multilingual",
"neural",
"machine",
"translation"
],
"offsets... | [
"although",
"multilingual",
"neural",
"machine",
"translation",
"(",
"mnmt",
")",
"enables",
"multiple",
"language",
"translations",
",",
"the",
"training",
"process",
"is",
"based",
"on",
"independent",
"multilingual",
"objectives",
".",
"most",
"multilingual",
"mo... |
ACL | HiddenCut: Simple Data Augmentation for Natural Language Understanding with Better Generalizability | Fine-tuning large pre-trained models with task-specific data has achieved great success in NLP. However, it has been demonstrated that the majority of information within the self-attention networks is redundant and not utilized effectively during the fine-tuning stage. This leads to inferior results when generalizing t... | 551e6aa674c6e65e4d0771adf28ac803 | 2,021 | [
"fine - tuning large pre - trained models with task - specific data has achieved great success in nlp .",
"however , it has been demonstrated that the majority of information within the self - attention networks is redundant and not utilized effectively during the fine - tuning stage .",
"this leads to inferior... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "nlp",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"nlp"
],
"offsets": [
18
]
}
],
"trigger": {
"text": "achieved",
"tokens": [
... | [
"fine",
"-",
"tuning",
"large",
"pre",
"-",
"trained",
"models",
"with",
"task",
"-",
"specific",
"data",
"has",
"achieved",
"great",
"success",
"in",
"nlp",
".",
"however",
",",
"it",
"has",
"been",
"demonstrated",
"that",
"the",
"majority",
"of",
"inform... |
ACL | Argument Invention from First Principles | Competitive debaters often find themselves facing a challenging task – how to debate a topic they know very little about, with only minutes to prepare, and without access to books or the Internet? What they often do is rely on ”first principles”, commonplace arguments which are relevant to many topics, and which they h... | a7c1cf4b0bb6f059db5dc97edec10c52 | 2,019 | [
"competitive debaters often find themselves facing a challenging task – how to debate a topic they know very little about , with only minutes to prepare , and without access to books or the internet ?",
"what they often do is rely on ” first principles ” , commonplace arguments which are relevant to many topics ,... | [
{
"event_type": "WKS",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Researcher",
"tokens": [
"we"
],
"offsets": [
69
]
},
{
"text": "these arguments",
"nugget_type": "TAK",
... | [
"competitive",
"debaters",
"often",
"find",
"themselves",
"facing",
"a",
"challenging",
"task",
"–",
"how",
"to",
"debate",
"a",
"topic",
"they",
"know",
"very",
"little",
"about",
",",
"with",
"only",
"minutes",
"to",
"prepare",
",",
"and",
"without",
"acce... |
ACL | Semi-Supervised Dialogue Policy Learning via Stochastic Reward Estimation | Dialogue policy optimization often obtains feedback until task completion in task-oriented dialogue systems. This is insufficient for training intermediate dialogue turns since supervision signals (or rewards) are only provided at the end of dialogues. To address this issue, reward learning has been introduced to learn... | b81853827d2e80d49644e9625fa3c1a2 | 2,020 | [
"dialogue policy optimization often obtains feedback until task completion in task - oriented dialogue systems .",
"this is insufficient for training intermediate dialogue turns since supervision signals ( or rewards ) are only provided at the end of dialogues .",
"to address this issue , reward learning has be... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "task - oriented dialogue systems",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"task",
"-",
"oriented",
"dialogue",
"systems"
],
"offsets"... | [
"dialogue",
"policy",
"optimization",
"often",
"obtains",
"feedback",
"until",
"task",
"completion",
"in",
"task",
"-",
"oriented",
"dialogue",
"systems",
".",
"this",
"is",
"insufficient",
"for",
"training",
"intermediate",
"dialogue",
"turns",
"since",
"supervisio... |
ACL | ESPRIT: Explaining Solutions to Physical Reasoning Tasks | Neural networks lack the ability to reason about qualitative physics and so cannot generalize to scenarios and tasks unseen during training. We propose ESPRIT, a framework for commonsense reasoning about qualitative physics in natural language that generates interpretable descriptions of physical events. We use a two-s... | e976c7ef98fe6e7ac732d5298b41ed53 | 2,020 | [
"neural networks lack the ability to reason about qualitative physics and so cannot generalize to scenarios and tasks unseen during training .",
"we propose esprit , a framework for commonsense reasoning about qualitative physics in natural language that generates interpretable descriptions of physical events .",... | [
{
"event_type": "RWF",
"arguments": [
{
"text": "neural networks",
"nugget_type": "APP",
"argument_type": "Concern",
"tokens": [
"neural",
"networks"
],
"offsets": [
0,
1
]
},
{
"text": ... | [
"neural",
"networks",
"lack",
"the",
"ability",
"to",
"reason",
"about",
"qualitative",
"physics",
"and",
"so",
"cannot",
"generalize",
"to",
"scenarios",
"and",
"tasks",
"unseen",
"during",
"training",
".",
"we",
"propose",
"esprit",
",",
"a",
"framework",
"f... |
ACL | Deep Dominance - How to Properly Compare Deep Neural Models | Comparing between Deep Neural Network (DNN) models based on their performance on unseen data is crucial for the progress of the NLP field. However, these models have a large number of hyper-parameters and, being non-convex, their convergence point depends on the random values chosen at initialization and during trainin... | caa5b388eefe9ef59c000b5c8797c1fc | 2,019 | [
"comparing between deep neural network ( dnn ) models based on their performance on unseen data is crucial for the progress of the nlp field .",
"however , these models have a large number of hyper - parameters and , being non - convex , their convergence point depends on the random values chosen at initializatio... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "deep neural network ( dnn ) models",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"deep",
"neural",
"network",
"(",
"dnn",
")",
"models... | [
"comparing",
"between",
"deep",
"neural",
"network",
"(",
"dnn",
")",
"models",
"based",
"on",
"their",
"performance",
"on",
"unseen",
"data",
"is",
"crucial",
"for",
"the",
"progress",
"of",
"the",
"nlp",
"field",
".",
"however",
",",
"these",
"models",
"... |
ACL | An Analysis of Negation in Natural Language Understanding Corpora | This paper analyzes negation in eight popular corpora spanning six natural language understanding tasks. We show that these corpora have few negations compared to general-purpose English, and that the few negations in them are often unimportant. Indeed, one can often ignore negations and still make the right prediction... | 19a3af54c0756f22a5c511695d3f95ae | 2,022 | [
"this paper analyzes negation in eight popular corpora spanning six natural language understanding tasks .",
"we show that these corpora have few negations compared to general - purpose english , and that the few negations in them are often unimportant .",
"indeed , one can often ignore negations and still make... | [
{
"event_type": "FIN",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Finder",
"tokens": [
"we"
],
"offsets": [
15
]
},
{
"text": "have",
"nugget_type": "E-CMP",
"argu... | [
"this",
"paper",
"analyzes",
"negation",
"in",
"eight",
"popular",
"corpora",
"spanning",
"six",
"natural",
"language",
"understanding",
"tasks",
".",
"we",
"show",
"that",
"these",
"corpora",
"have",
"few",
"negations",
"compared",
"to",
"general",
"-",
"purpos... |
ACL | Diversifying Dialogue Generation with Non-Conversational Text | Neural network-based sequence-to-sequence (seq2seq) models strongly suffer from the low-diversity problem when it comes to open-domain dialogue generation. As bland and generic utterances usually dominate the frequency distribution in our daily chitchat, avoiding them to generate more interesting responses requires com... | f212a4848104b17f1c6f9d8dada351f6 | 2,020 | [
"neural network - based sequence - to - sequence ( seq2seq ) models strongly suffer from the low - diversity problem when it comes to open - domain dialogue generation .",
"as bland and generic utterances usually dominate the frequency distribution in our daily chitchat , avoiding them to generate more interestin... | [
{
"event_type": "RWF",
"arguments": [
{
"text": "neural network - based sequence - to - sequence models",
"nugget_type": "APP",
"argument_type": "Concern",
"tokens": [
"neural",
"network",
"-",
"based",
"sequence",
... | [
"neural",
"network",
"-",
"based",
"sequence",
"-",
"to",
"-",
"sequence",
"(",
"seq2seq",
")",
"models",
"strongly",
"suffer",
"from",
"the",
"low",
"-",
"diversity",
"problem",
"when",
"it",
"comes",
"to",
"open",
"-",
"domain",
"dialogue",
"generation",
... |
ACL | DoQA - Accessing Domain-Specific FAQs via Conversational QA | The goal of this work is to build conversational Question Answering (QA) interfaces for the large body of domain-specific information available in FAQ sites. We present DoQA, a dataset with 2,437 dialogues and 10,917 QA pairs. The dialogues are collected from three Stack Exchange sites using the Wizard of Oz method wit... | 20908bb13ffb61303710bf08e92799ca | 2,020 | [
"the goal of this work is to build conversational question answering ( qa ) interfaces for the large body of domain - specific information available in faq sites .",
"we present doqa , a dataset with 2 , 437 dialogues and 10 , 917 qa pairs .",
"the dialogues are collected from three stack exchange sites using t... | [
{
"event_type": "WKS",
"arguments": [
{
"text": "conversational question answering interfaces",
"nugget_type": "TAK",
"argument_type": "Content",
"tokens": [
"conversational",
"question",
"answering",
"interfaces"
],
... | [
"the",
"goal",
"of",
"this",
"work",
"is",
"to",
"build",
"conversational",
"question",
"answering",
"(",
"qa",
")",
"interfaces",
"for",
"the",
"large",
"body",
"of",
"domain",
"-",
"specific",
"information",
"available",
"in",
"faq",
"sites",
".",
"we",
... |
ACL | Simple and Effective Knowledge-Driven Query Expansion for QA-Based Product Attribute Extraction | A key challenge in attribute value extraction (AVE) from e-commerce sites is how to handle a large number of attributes for diverse products. Although this challenge is partially addressed by a question answering (QA) approach which finds a value in product data for a given query (attribute), it does not work effective... | 822657195394883e7fedb7b0afb5f918 | 2,022 | [
"a key challenge in attribute value extraction ( ave ) from e - commerce sites is how to handle a large number of attributes for diverse products .",
"although this challenge is partially addressed by a question answering ( qa ) approach which finds a value in product data for a given query ( attribute ) , it doe... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "attribute value extraction",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"attribute",
"value",
"extraction"
],
"offsets": [
4,
5,
... | [
"a",
"key",
"challenge",
"in",
"attribute",
"value",
"extraction",
"(",
"ave",
")",
"from",
"e",
"-",
"commerce",
"sites",
"is",
"how",
"to",
"handle",
"a",
"large",
"number",
"of",
"attributes",
"for",
"diverse",
"products",
".",
"although",
"this",
"chal... |
ACL | You Impress Me: Dialogue Generation via Mutual Persona Perception | Despite the continuing efforts to improve the engagingness and consistency of chit-chat dialogue systems, the majority of current work simply focus on mimicking human-like responses, leaving understudied the aspects of modeling understanding between interlocutors. The research in cognitive science, instead, suggests th... | 2771f84a80bf7e58edea17b52a9a7952 | 2,020 | [
"despite the continuing efforts to improve the engagingness and consistency of chit - chat dialogue systems , the majority of current work simply focus on mimicking human - like responses , leaving understudied the aspects of modeling understanding between interlocutors .",
"the research in cognitive science , in... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "chit - chat dialogue systems",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"chit",
"-",
"chat",
"dialogue",
"systems"
],
"offsets": [
... | [
"despite",
"the",
"continuing",
"efforts",
"to",
"improve",
"the",
"engagingness",
"and",
"consistency",
"of",
"chit",
"-",
"chat",
"dialogue",
"systems",
",",
"the",
"majority",
"of",
"current",
"work",
"simply",
"focus",
"on",
"mimicking",
"human",
"-",
"lik... |
ACL | PLATO: Pre-trained Dialogue Generation Model with Discrete Latent Variable | Pre-training models have been proved effective for a wide range of natural language processing tasks. Inspired by this, we propose a novel dialogue generation pre-training framework to support various kinds of conversations, including chit-chat, knowledge grounded dialogues, and conversational question answering. In th... | 350154ab1dcf6b0aa4766d1cdd133453 | 2,020 | [
"pre - training models have been proved effective for a wide range of natural language processing tasks .",
"inspired by this , we propose a novel dialogue generation pre - training framework to support various kinds of conversations , including chit - chat , knowledge grounded dialogues , and conversational ques... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "pre - training models",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"pre",
"-",
"training",
"models"
],
"offsets": [
0,
1,
... | [
"pre",
"-",
"training",
"models",
"have",
"been",
"proved",
"effective",
"for",
"a",
"wide",
"range",
"of",
"natural",
"language",
"processing",
"tasks",
".",
"inspired",
"by",
"this",
",",
"we",
"propose",
"a",
"novel",
"dialogue",
"generation",
"pre",
"-",... |
ACL | Improving Chinese Word Segmentation with Wordhood Memory Networks | Contextual features always play an important role in Chinese word segmentation (CWS). Wordhood information, being one of the contextual features, is proved to be useful in many conventional character-based segmenters. However, this feature receives less attention in recent neural models and it is also challenging to de... | aaebc5ee996d5628530d2e8686f4692b | 2,020 | [
"contextual features always play an important role in chinese word segmentation ( cws ) .",
"wordhood information , being one of the contextual features , is proved to be useful in many conventional character - based segmenters .",
"however , this feature receives less attention in recent neural models and it i... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "chinese word segmentation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"chinese",
"word",
"segmentation"
],
"offsets": [
8,
9,
... | [
"contextual",
"features",
"always",
"play",
"an",
"important",
"role",
"in",
"chinese",
"word",
"segmentation",
"(",
"cws",
")",
".",
"wordhood",
"information",
",",
"being",
"one",
"of",
"the",
"contextual",
"features",
",",
"is",
"proved",
"to",
"be",
"use... |
ACL | A Joint Model for Dropped Pronoun Recovery and Conversational Discourse Parsing in Chinese Conversational Speech | In this paper, we present a neural model for joint dropped pronoun recovery (DPR) and conversational discourse parsing (CDP) in Chinese conversational speech. We show that DPR and CDP are closely related, and a joint model benefits both tasks. We refer to our model as DiscProReco, and it first encodes the tokens in eac... | 2ce38c653341c2c99e7ffe09db051f40 | 2,021 | [
"in this paper , we present a neural model for joint dropped pronoun recovery ( dpr ) and conversational discourse parsing ( cdp ) in chinese conversational speech .",
"we show that dpr and cdp are closely related , and a joint model benefits both tasks .",
"we refer to our model as discproreco , and it first e... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
4
]
},
{
"text": "neural model",
"nugget_type": "APP",
... | [
"in",
"this",
"paper",
",",
"we",
"present",
"a",
"neural",
"model",
"for",
"joint",
"dropped",
"pronoun",
"recovery",
"(",
"dpr",
")",
"and",
"conversational",
"discourse",
"parsing",
"(",
"cdp",
")",
"in",
"chinese",
"conversational",
"speech",
".",
"we",
... |
ACL | DeepRapper: Neural Rap Generation with Rhyme and Rhythm Modeling | Rap generation, which aims to produce lyrics and corresponding singing beats, needs to model both rhymes and rhythms. Previous works for rap generation focused on rhyming lyrics, but ignored rhythmic beats, which are important for rap performance. In this paper, we develop DeepRapper, a Transformer-based rap generation... | 1184d734c6d5c9d22f5268f313c8306f | 2,021 | [
"rap generation , which aims to produce lyrics and corresponding singing beats , needs to model both rhymes and rhythms .",
"previous works for rap generation focused on rhyming lyrics , but ignored rhythmic beats , which are important for rap performance .",
"in this paper , we develop deeprapper , a transform... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "rap generation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"rap",
"generation"
],
"offsets": [
0,
1
]
}
],
"trigger": {
... | [
"rap",
"generation",
",",
"which",
"aims",
"to",
"produce",
"lyrics",
"and",
"corresponding",
"singing",
"beats",
",",
"needs",
"to",
"model",
"both",
"rhymes",
"and",
"rhythms",
".",
"previous",
"works",
"for",
"rap",
"generation",
"focused",
"on",
"rhyming",... |
ACL | ARNOR: Attention Regularization based Noise Reduction for Distant Supervision Relation Classification | Distant supervision is widely used in relation classification in order to create large-scale training data by aligning a knowledge base with an unlabeled corpus. However, it also introduces amounts of noisy labels where a contextual sentence actually does not express the labeled relation. In this paper, we propose ARNO... | 55e42c83b8f4f408326e55df0d884e52 | 2,019 | [
"distant supervision is widely used in relation classification in order to create large - scale training data by aligning a knowledge base with an unlabeled corpus .",
"however , it also introduces amounts of noisy labels where a contextual sentence actually does not express the labeled relation .",
"in this pa... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "distant supervision",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"distant",
"supervision"
],
"offsets": [
0,
1
]
}
],
"trigge... | [
"distant",
"supervision",
"is",
"widely",
"used",
"in",
"relation",
"classification",
"in",
"order",
"to",
"create",
"large",
"-",
"scale",
"training",
"data",
"by",
"aligning",
"a",
"knowledge",
"base",
"with",
"an",
"unlabeled",
"corpus",
".",
"however",
","... |
ACL | EntSUM: A Data Set for Entity-Centric Extractive Summarization | Controllable summarization aims to provide summaries that take into account user-specified aspects and preferences to better assist them with their information need, as opposed to the standard summarization setup which build a single generic summary of a document.We introduce a human-annotated data set EntSUM for contr... | 023fdf1026d094faa4a632135fcec9b9 | 2,022 | [
"controllable summarization aims to provide summaries that take into account user - specified aspects and preferences to better assist them with their information need , as opposed to the standard summarization setup which build a single generic summary of a document .",
"we introduce a human - annotated data set... | [
{
"event_type": "RWS",
"arguments": [
{
"text": "controllable summarization",
"nugget_type": "APP",
"argument_type": "Subject",
"tokens": [
"controllable",
"summarization"
],
"offsets": [
0,
1
]
},
... | [
"controllable",
"summarization",
"aims",
"to",
"provide",
"summaries",
"that",
"take",
"into",
"account",
"user",
"-",
"specified",
"aspects",
"and",
"preferences",
"to",
"better",
"assist",
"them",
"with",
"their",
"information",
"need",
",",
"as",
"opposed",
"... |
ACL | Span Selection Pre-training for Question Answering | BERT (Bidirectional Encoder Representations from Transformers) and related pre-trained Transformers have provided large gains across many language understanding tasks, achieving a new state-of-the-art (SOTA). BERT is pretrained on two auxiliary tasks: Masked Language Model and Next Sentence Prediction. In this paper we... | e68bb3f8f1b477afe05e48ba9b22b1a5 | 2,020 | [
"bert ( bidirectional encoder representations from transformers ) and related pre - trained transformers have provided large gains across many language understanding tasks , achieving a new state - of - the - art ( sota ) .",
"bert is pretrained on two auxiliary tasks : masked language model and next sentence pre... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "bert",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"bert"
],
"offsets": [
0
]
}
],
"trigger": {
"text": "provided",
"tokens": [
... | [
"bert",
"(",
"bidirectional",
"encoder",
"representations",
"from",
"transformers",
")",
"and",
"related",
"pre",
"-",
"trained",
"transformers",
"have",
"provided",
"large",
"gains",
"across",
"many",
"language",
"understanding",
"tasks",
",",
"achieving",
"a",
"... |
ACL | Curriculum Pre-training for End-to-End Speech Translation | End-to-end speech translation poses a heavy burden on the encoder because it has to transcribe, understand, and learn cross-lingual semantics simultaneously. To obtain a powerful encoder, traditional methods pre-train it on ASR data to capture speech features. However, we argue that pre-training the encoder only throug... | 046bbb1a05879e77b13b105e45c8a4b1 | 2,020 | [
"end - to - end speech translation poses a heavy burden on the encoder because it has to transcribe , understand , and learn cross - lingual semantics simultaneously .",
"to obtain a powerful encoder , traditional methods pre - train it on asr data to capture speech features .",
"however , we argue that pre - t... | [
{
"event_type": "RWF",
"arguments": [
{
"text": "end - to - end speech translation",
"nugget_type": "APP",
"argument_type": "Concern",
"tokens": [
"end",
"-",
"to",
"-",
"end",
"speech",
"translation"... | [
"end",
"-",
"to",
"-",
"end",
"speech",
"translation",
"poses",
"a",
"heavy",
"burden",
"on",
"the",
"encoder",
"because",
"it",
"has",
"to",
"transcribe",
",",
"understand",
",",
"and",
"learn",
"cross",
"-",
"lingual",
"semantics",
"simultaneously",
".",
... |
ACL | Probabilistic, Structure-Aware Algorithms for Improved Variety, Accuracy, and Coverage of AMR Alignments | We present algorithms for aligning components of Abstract Meaning Representation (AMR) graphs to spans in English sentences. We leverage unsupervised learning in combination with heuristics, taking the best of both worlds from previous AMR aligners. Our unsupervised models, however, are more sensitive to graph substruc... | 8e8459f99d1cd027d2b2809ebe646229 | 2,021 | [
"we present algorithms for aligning components of abstract meaning representation ( amr ) graphs to spans in english sentences .",
"we leverage unsupervised learning in combination with heuristics , taking the best of both worlds from previous amr aligners .",
"our unsupervised models , however , are more sensi... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
0
]
},
{
"text": "algorithms",
"nugget_type": "APP",
... | [
"we",
"present",
"algorithms",
"for",
"aligning",
"components",
"of",
"abstract",
"meaning",
"representation",
"(",
"amr",
")",
"graphs",
"to",
"spans",
"in",
"english",
"sentences",
".",
"we",
"leverage",
"unsupervised",
"learning",
"in",
"combination",
"with",
... |
ACL | Measuring and Increasing Context Usage in Context-Aware Machine Translation | Recent work in neural machine translation has demonstrated both the necessity and feasibility of using inter-sentential context, context from sentences other than those currently being translated. However, while many current methods present model architectures that theoretically can use this extra context, it is often ... | 62bd7b7c9817f82f0a0d3e5393a881a8 | 2,021 | [
"recent work in neural machine translation has demonstrated both the necessity and feasibility of using inter - sentential context , context from sentences other than those currently being translated .",
"however , while many current methods present model architectures that theoretically can use this extra contex... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "neural machine translation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"neural",
"machine",
"translation"
],
"offsets": [
3,
4,
... | [
"recent",
"work",
"in",
"neural",
"machine",
"translation",
"has",
"demonstrated",
"both",
"the",
"necessity",
"and",
"feasibility",
"of",
"using",
"inter",
"-",
"sentential",
"context",
",",
"context",
"from",
"sentences",
"other",
"than",
"those",
"currently",
... |
ACL | Hooks in the Headline: Learning to Generate Headlines with Controlled Styles | Current summarization systems only produce plain, factual headlines, far from the practical needs for the exposure and memorableness of the articles. We propose a new task, Stylistic Headline Generation (SHG), to enrich the headlines with three style options (humor, romance and clickbait), thus attracting more readers.... | 4a9ab0c72294b4adaaa8ef571dfae5b4 | 2,020 | [
"current summarization systems only produce plain , factual headlines , far from the practical needs for the exposure and memorableness of the articles .",
"we propose a new task , stylistic headline generation ( shg ) , to enrich the headlines with three style options ( humor , romance and clickbait ) , thus att... | [
{
"event_type": "RWF",
"arguments": [
{
"text": "current summarization systems",
"nugget_type": "APP",
"argument_type": "Concern",
"tokens": [
"current",
"summarization",
"systems"
],
"offsets": [
0,
1,
... | [
"current",
"summarization",
"systems",
"only",
"produce",
"plain",
",",
"factual",
"headlines",
",",
"far",
"from",
"the",
"practical",
"needs",
"for",
"the",
"exposure",
"and",
"memorableness",
"of",
"the",
"articles",
".",
"we",
"propose",
"a",
"new",
"task"... |
ACL | Have my arguments been replied to? Argument Pair Extraction as Machine Reading Comprehension | Argument pair extraction (APE) aims to automatically mine argument pairs from two interrelated argumentative documents. Existing studies typically identify argument pairs indirectly by predicting sentence-level relations between two documents, neglecting the modeling of the holistic argument-level interactions. Towards... | b635e461bf33e7a017c173ce43254ae9 | 2,022 | [
"argument pair extraction ( ape ) aims to automatically mine argument pairs from two interrelated argumentative documents .",
"existing studies typically identify argument pairs indirectly by predicting sentence - level relations between two documents , neglecting the modeling of the holistic argument - level int... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "argument pair extraction",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"argument",
"pair",
"extraction"
],
"offsets": [
0,
1,
2
... | [
"argument",
"pair",
"extraction",
"(",
"ape",
")",
"aims",
"to",
"automatically",
"mine",
"argument",
"pairs",
"from",
"two",
"interrelated",
"argumentative",
"documents",
".",
"existing",
"studies",
"typically",
"identify",
"argument",
"pairs",
"indirectly",
"by",
... |
ACL | Compositionality and Generalization In Emergent Languages | Natural language allows us to refer to novel composite concepts by combining expressions denoting their parts according to systematic rules, a property known as compositionality. In this paper, we study whether the language emerging in deep multi-agent simulations possesses a similar ability to refer to novel primitive... | 03314ef93816ed51959399f0216f2eab | 2,020 | [
"natural language allows us to refer to novel composite concepts by combining expressions denoting their parts according to systematic rules , a property known as compositionality .",
"in this paper , we study whether the language emerging in deep multi - agent simulations possesses a similar ability to refer to ... | [
{
"event_type": "WKS",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Researcher",
"tokens": [
"we"
],
"offsets": [
31
]
},
{
"text": "language emerging",
"nugget_type": "TAK"... | [
"natural",
"language",
"allows",
"us",
"to",
"refer",
"to",
"novel",
"composite",
"concepts",
"by",
"combining",
"expressions",
"denoting",
"their",
"parts",
"according",
"to",
"systematic",
"rules",
",",
"a",
"property",
"known",
"as",
"compositionality",
".",
... |
ACL | MSCTD: A Multimodal Sentiment Chat Translation Dataset | Multimodal machine translation and textual chat translation have received considerable attention in recent years. Although the conversation in its natural form is usually multimodal, there still lacks work on multimodal machine translation in conversations. In this work, we introduce a new task named Multimodal Chat Tr... | 279762b99de7c5ba979e14acaf883c76 | 2,022 | [
"multimodal machine translation and textual chat translation have received considerable attention in recent years .",
"although the conversation in its natural form is usually multimodal , there still lacks work on multimodal machine translation in conversations .",
"in this work , we introduce a new task named... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "multimodal machine translation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"multimodal",
"machine",
"translation"
],
"offsets": [
0,
1,
... | [
"multimodal",
"machine",
"translation",
"and",
"textual",
"chat",
"translation",
"have",
"received",
"considerable",
"attention",
"in",
"recent",
"years",
".",
"although",
"the",
"conversation",
"in",
"its",
"natural",
"form",
"is",
"usually",
"multimodal",
",",
"... |
ACL | AligNarr: Aligning Narratives on Movies | High-quality alignment between movie scripts and plot summaries is an asset for learning to summarize stories and to generate dialogues. The alignment task is challenging as scripts and summaries substantially differ in details and abstraction levels as well as in linguistic register. This paper addresses the alignment... | 50c715a946e46f25ae6bf254babd77f8 | 2,021 | [
"high - quality alignment between movie scripts and plot summaries is an asset for learning to summarize stories and to generate dialogues .",
"the alignment task is challenging as scripts and summaries substantially differ in details and abstraction levels as well as in linguistic register .",
"this paper addr... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "high - quality alignment",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"high",
"-",
"quality",
"alignment"
],
"offsets": [
0,
1,... | [
"high",
"-",
"quality",
"alignment",
"between",
"movie",
"scripts",
"and",
"plot",
"summaries",
"is",
"an",
"asset",
"for",
"learning",
"to",
"summarize",
"stories",
"and",
"to",
"generate",
"dialogues",
".",
"the",
"alignment",
"task",
"is",
"challenging",
"a... |
ACL | Towards Faithfully Interpretable NLP Systems: How Should We Define and Evaluate Faithfulness? | With the growing popularity of deep-learning based NLP models, comes a need for interpretable systems. But what is interpretability, and what constitutes a high-quality interpretation? In this opinion piece we reflect on the current state of interpretability evaluation research. We call for more clearly differentiating... | 5c89fcb746bed1692d6cec80df59f613 | 2,020 | [
"with the growing popularity of deep - learning based nlp models , comes a need for interpretable systems .",
"but what is interpretability , and what constitutes a high - quality interpretation ?",
"in this opinion piece we reflect on the current state of interpretability evaluation research .",
"we call for... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "interpretability evaluation research",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"interpretability",
"evaluation",
"research"
],
"offsets": [
44,
... | [
"with",
"the",
"growing",
"popularity",
"of",
"deep",
"-",
"learning",
"based",
"nlp",
"models",
",",
"comes",
"a",
"need",
"for",
"interpretable",
"systems",
".",
"but",
"what",
"is",
"interpretability",
",",
"and",
"what",
"constitutes",
"a",
"high",
"-",
... |
ACL | k-Rater Reliability: The Correct Unit of Reliability for Aggregated Human Annotations | Since the inception of crowdsourcing, aggregation has been a common strategy for dealing with unreliable data. Aggregate ratings are more reliable than individual ones. However, many Natural Language Processing (NLP) applications that rely on aggregate ratings only report the reliability of individual ratings, which is... | 441c886efb209902d21ad730fa7997c2 | 2,022 | [
"since the inception of crowdsourcing , aggregation has been a common strategy for dealing with unreliable data .",
"aggregate ratings are more reliable than individual ones .",
"however , many natural language processing ( nlp ) applications that rely on aggregate ratings only report the reliability of individ... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "aggregation",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"aggregation"
],
"offsets": [
6
]
}
],
"trigger": {
"text": "strategy",
"tok... | [
"since",
"the",
"inception",
"of",
"crowdsourcing",
",",
"aggregation",
"has",
"been",
"a",
"common",
"strategy",
"for",
"dealing",
"with",
"unreliable",
"data",
".",
"aggregate",
"ratings",
"are",
"more",
"reliable",
"than",
"individual",
"ones",
".",
"however"... |
ACL | Directed Acyclic Graph Network for Conversational Emotion Recognition | The modeling of conversational context plays a vital role in emotion recognition from conversation (ERC). In this paper, we put forward a novel idea of encoding the utterances with a directed acyclic graph (DAG) to better model the intrinsic structure within a conversation, and design a directed acyclic neural network,... | 702e58c75bac4615661db67dab2bd2ea | 2,021 | [
"the modeling of conversational context plays a vital role in emotion recognition from conversation ( erc ) .",
"in this paper , we put forward a novel idea of encoding the utterances with a directed acyclic graph ( dag ) to better model the intrinsic structure within a conversation , and design a directed acycli... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "emotion recognition from conversation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"emotion",
"recognition",
"from",
"conversation"
],
"offsets": [... | [
"the",
"modeling",
"of",
"conversational",
"context",
"plays",
"a",
"vital",
"role",
"in",
"emotion",
"recognition",
"from",
"conversation",
"(",
"erc",
")",
".",
"in",
"this",
"paper",
",",
"we",
"put",
"forward",
"a",
"novel",
"idea",
"of",
"encoding",
"... |
ACL | On Compositional Generalization of Neural Machine Translation | Modern neural machine translation (NMT) models have achieved competitive performance in standard benchmarks such as WMT. However, there still exist significant issues such as robustness, domain generalization, etc. In this paper, we study NMT models from the perspective of compositional generalization by building a ben... | 7fe816b9e10a0f8ada81442e9e2b2bcd | 2,021 | [
"modern neural machine translation ( nmt ) models have achieved competitive performance in standard benchmarks such as wmt .",
"however , there still exist significant issues such as robustness , domain generalization , etc .",
"in this paper , we study nmt models from the perspective of compositional generaliz... | [
{
"event_type": "FAC",
"arguments": [
{
"text": "nmt model",
"nugget_type": "APP",
"argument_type": "Subject",
"tokens": [
"nmt",
"model"
],
"offsets": [
83,
84
]
},
{
"text": "compositi... | [
"modern",
"neural",
"machine",
"translation",
"(",
"nmt",
")",
"models",
"have",
"achieved",
"competitive",
"performance",
"in",
"standard",
"benchmarks",
"such",
"as",
"wmt",
".",
"however",
",",
"there",
"still",
"exist",
"significant",
"issues",
"such",
"as",... |
ACL | VECO: Variable and Flexible Cross-lingual Pre-training for Language Understanding and Generation | Existing work in multilingual pretraining has demonstrated the potential of cross-lingual transferability by training a unified Transformer encoder for multiple languages. However, much of this work only relies on the shared vocabulary and bilingual contexts to encourage the correlation across languages, which is loose... | 990b4b4c629b8e166c2aa95bfcf59c00 | 2,021 | [
"existing work in multilingual pretraining has demonstrated the potential of cross - lingual transferability by training a unified transformer encoder for multiple languages .",
"however , much of this work only relies on the shared vocabulary and bilingual contexts to encourage the correlation across languages ,... | [
{
"event_type": "RWS",
"arguments": [
{
"text": "existing work",
"nugget_type": "APP",
"argument_type": "Subject",
"tokens": [
"existing",
"work"
],
"offsets": [
0,
1
]
},
{
"text": "in ... | [
"existing",
"work",
"in",
"multilingual",
"pretraining",
"has",
"demonstrated",
"the",
"potential",
"of",
"cross",
"-",
"lingual",
"transferability",
"by",
"training",
"a",
"unified",
"transformer",
"encoder",
"for",
"multiple",
"languages",
".",
"however",
",",
"... |
ACL | Modeling Task-Aware MIMO Cardinality for Efficient Multilingual Neural Machine Translation | Neural machine translation has achieved great success in bilingual settings, as well as in multilingual settings. With the increase of the number of languages, multilingual systems tend to underperform their bilingual counterparts. Model capacity has been found crucial for massively multilingual NMT to support language... | 3f2aaa6fd508313022c648d4b4ba6807 | 2,021 | [
"neural machine translation has achieved great success in bilingual settings , as well as in multilingual settings .",
"with the increase of the number of languages , multilingual systems tend to underperform their bilingual counterparts .",
"model capacity has been found crucial for massively multilingual nmt ... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "neural machine translation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"neural",
"machine",
"translation"
],
"offsets": [
0,
1,
... | [
"neural",
"machine",
"translation",
"has",
"achieved",
"great",
"success",
"in",
"bilingual",
"settings",
",",
"as",
"well",
"as",
"in",
"multilingual",
"settings",
".",
"with",
"the",
"increase",
"of",
"the",
"number",
"of",
"languages",
",",
"multilingual",
... |
ACL | Deep Unknown Intent Detection with Margin Loss | Identifying the unknown (novel) user intents that have never appeared in the training set is a challenging task in the dialogue system. In this paper, we present a two-stage method for detecting unknown intents. We use bidirectional long short-term memory (BiLSTM) network with the margin loss as the feature extractor. ... | 750b83fdaa7c255b56b24a6edbd9f906 | 2,019 | [
"identifying the unknown ( novel ) user intents that have never appeared in the training set is a challenging task in the dialogue system .",
"in this paper , we present a two - stage method for detecting unknown intents .",
"we use bidirectional long short - term memory ( bilstm ) network with the margin loss ... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "user intents",
"nugget_type": "FEA",
"argument_type": "Target",
"tokens": [
"user",
"intents"
],
"offsets": [
6,
7
]
}
],
"trigger": {
"t... | [
"identifying",
"the",
"unknown",
"(",
"novel",
")",
"user",
"intents",
"that",
"have",
"never",
"appeared",
"in",
"the",
"training",
"set",
"is",
"a",
"challenging",
"task",
"in",
"the",
"dialogue",
"system",
".",
"in",
"this",
"paper",
",",
"we",
"present... |
ACL | Redistributing Low-Frequency Words: Making the Most of Monolingual Data in Non-Autoregressive Translation | Knowledge distillation (KD) is the preliminary step for training non-autoregressive translation (NAT) models, which eases the training of NAT models at the cost of losing important information for translating low-frequency words. In this work, we provide an appealing alternative for NAT – monolingual KD, which trains N... | 61587da2756d4dcc5a20db6d083e59f4 | 2,022 | [
"knowledge distillation ( kd ) is the preliminary step for training non - autoregressive translation ( nat ) models , which eases the training of nat models at the cost of losing important information for translating low - frequency words .",
"in this work , we provide an appealing alternative for nat – monolingu... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "nat models",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"nat",
"models"
],
"offsets": [
51,
18
]
}
],
"trigger": {
"tex... | [
"knowledge",
"distillation",
"(",
"kd",
")",
"is",
"the",
"preliminary",
"step",
"for",
"training",
"non",
"-",
"autoregressive",
"translation",
"(",
"nat",
")",
"models",
",",
"which",
"eases",
"the",
"training",
"of",
"nat",
"models",
"at",
"the",
"cost",
... |
ACL | Sentence Meta-Embeddings for Unsupervised Semantic Textual Similarity | We address the task of unsupervised Semantic Textual Similarity (STS) by ensembling diverse pre-trained sentence encoders into sentence meta-embeddings. We apply, extend and evaluate different meta-embedding methods from the word embedding literature at the sentence level, including dimensionality reduction (Yin and Sc... | e8222042d50267242a226e976fc22f6d | 2,020 | [
"we address the task of unsupervised semantic textual similarity ( sts ) by ensembling diverse pre - trained sentence encoders into sentence meta - embeddings .",
"we apply , extend and evaluate different meta - embedding methods from the word embedding literature at the sentence level , including dimensionality ... | [
{
"event_type": "MDS",
"arguments": [
{
"text": "diverse pre - trained sentence encoders",
"nugget_type": "MOD",
"argument_type": "BaseComponent",
"tokens": [
"diverse",
"pre",
"-",
"trained",
"sentence",
"enco... | [
"we",
"address",
"the",
"task",
"of",
"unsupervised",
"semantic",
"textual",
"similarity",
"(",
"sts",
")",
"by",
"ensembling",
"diverse",
"pre",
"-",
"trained",
"sentence",
"encoders",
"into",
"sentence",
"meta",
"-",
"embeddings",
".",
"we",
"apply",
",",
... |
ACL | Large Dataset and Language Model Fun-Tuning for Humor Recognition | The task of humor recognition has attracted a lot of attention recently due to the urge to process large amounts of user-generated texts and rise of conversational agents. We collected a dataset of jokes and funny dialogues in Russian from various online resources and complemented them carefully with unfunny texts with... | ddf26f34c91b8bc6a3cf464c0f69813e | 2,019 | [
"the task of humor recognition has attracted a lot of attention recently due to the urge to process large amounts of user - generated texts and rise of conversational agents .",
"we collected a dataset of jokes and funny dialogues in russian from various online resources and complemented them carefully with unfun... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "humor recognition",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"humor",
"recognition"
],
"offsets": [
3,
4
]
}
],
"trigger": ... | [
"the",
"task",
"of",
"humor",
"recognition",
"has",
"attracted",
"a",
"lot",
"of",
"attention",
"recently",
"due",
"to",
"the",
"urge",
"to",
"process",
"large",
"amounts",
"of",
"user",
"-",
"generated",
"texts",
"and",
"rise",
"of",
"conversational",
"agen... |
ACL | Gated Embeddings in End-to-End Speech Recognition for Conversational-Context Fusion | We present a novel conversational-context aware end-to-end speech recognizer based on a gated neural network that incorporates conversational-context/word/speech embeddings. Unlike conventional speech recognition models, our model learns longer conversational-context information that spans across sentences and is conse... | 89982dae2a53a82c789d2d7ff87869a7 | 2,019 | [
"we present a novel conversational - context aware end - to - end speech recognizer based on a gated neural network that incorporates conversational - context / word / speech embeddings .",
"unlike conventional speech recognition models , our model learns longer conversational - context information that spans acr... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
0
]
},
{
"text": "conversational - context aware end - to - end speec... | [
"we",
"present",
"a",
"novel",
"conversational",
"-",
"context",
"aware",
"end",
"-",
"to",
"-",
"end",
"speech",
"recognizer",
"based",
"on",
"a",
"gated",
"neural",
"network",
"that",
"incorporates",
"conversational",
"-",
"context",
"/",
"word",
"/",
"spe... |
ACL | RST Discourse Parsing with Second-Stage EDU-Level Pre-training | Pre-trained language models (PLMs) have shown great potentials in natural language processing (NLP) including rhetorical structure theory (RST) discourse parsing.Current PLMs are obtained by sentence-level pre-training, which is different from the basic processing unit, i.e. element discourse unit (EDU).To this end, we... | 38ccc1d68cf7687a5776d7e0fe782e7a | 2,022 | [
"pre - trained language models ( plms ) have shown great potentials in natural language processing ( nlp ) including rhetorical structure theory ( rst ) discourse parsing .",
"current plms are obtained by sentence - level pre - training , which is different from the basic processing unit , i . e . element discour... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "natural language processing",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"natural",
"language",
"processing"
],
"offsets": [
13,
14,
... | [
"pre",
"-",
"trained",
"language",
"models",
"(",
"plms",
")",
"have",
"shown",
"great",
"potentials",
"in",
"natural",
"language",
"processing",
"(",
"nlp",
")",
"including",
"rhetorical",
"structure",
"theory",
"(",
"rst",
")",
"discourse",
"parsing",
".",
... |
ACL | Supporting Cognitive and Emotional Empathic Writing of Students | We present an annotation approach to capturing emotional and cognitive empathy in student-written peer reviews on business models in German. We propose an annotation scheme that allows us to model emotional and cognitive empathy scores based on three types of review components. Also, we conducted an annotation study wi... | 99ca7e30270e42180b3db2fd021a3064 | 2,021 | [
"we present an annotation approach to capturing emotional and cognitive empathy in student - written peer reviews on business models in german .",
"we propose an annotation scheme that allows us to model emotional and cognitive empathy scores based on three types of review components .",
"also , we conducted an... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
0
]
},
{
"text": "annotation approach",
"nugget_type": "APP",... | [
"we",
"present",
"an",
"annotation",
"approach",
"to",
"capturing",
"emotional",
"and",
"cognitive",
"empathy",
"in",
"student",
"-",
"written",
"peer",
"reviews",
"on",
"business",
"models",
"in",
"german",
".",
"we",
"propose",
"an",
"annotation",
"scheme",
... |
ACL | A Relaxed Matching Procedure for Unsupervised BLI | Recently unsupervised Bilingual Lexicon Induction(BLI) without any parallel corpus has attracted much research interest. One of the crucial parts in methods for the BLI task is the matching procedure. Previous works impose a too strong constraint on the matching and lead to many counterintuitive translation pairings. T... | ad5d97e9abab939ed7b6860afffa6647 | 2,020 | [
"recently unsupervised bilingual lexicon induction ( bli ) without any parallel corpus has attracted much research interest .",
"one of the crucial parts in methods for the bli task is the matching procedure .",
"previous works impose a too strong constraint on the matching and lead to many counterintuitive tra... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "unsupervised bilingual lexicon induction",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"unsupervised",
"bilingual",
"lexicon",
"induction"
],
"offse... | [
"recently",
"unsupervised",
"bilingual",
"lexicon",
"induction",
"(",
"bli",
")",
"without",
"any",
"parallel",
"corpus",
"has",
"attracted",
"much",
"research",
"interest",
".",
"one",
"of",
"the",
"crucial",
"parts",
"in",
"methods",
"for",
"the",
"bli",
"ta... |
ACL | Learning a Multi-Domain Curriculum for Neural Machine Translation | Most data selection research in machine translation focuses on improving a single domain. We perform data selection for multiple domains at once. This is achieved by carefully introducing instance-level domain-relevance features and automatically constructing a training curriculum to gradually concentrate on multi-doma... | a82203f064106d4c3ed5b6967dc90333 | 2,020 | [
"most data selection research in machine translation focuses on improving a single domain .",
"we perform data selection for multiple domains at once .",
"this is achieved by carefully introducing instance - level domain - relevance features and automatically constructing a training curriculum to gradually conc... | [
{
"event_type": "WKS",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Researcher",
"tokens": [
"we"
],
"offsets": [
14
]
},
{
"text": "data selection",
"nugget_type": "TAK",
... | [
"most",
"data",
"selection",
"research",
"in",
"machine",
"translation",
"focuses",
"on",
"improving",
"a",
"single",
"domain",
".",
"we",
"perform",
"data",
"selection",
"for",
"multiple",
"domains",
"at",
"once",
".",
"this",
"is",
"achieved",
"by",
"careful... |
ACL | Rewarding Smatch: Transition-Based AMR Parsing with Reinforcement Learning | Our work involves enriching the Stack-LSTM transition-based AMR parser (Ballesteros and Al-Onaizan, 2017) by augmenting training with Policy Learning and rewarding the Smatch score of sampled graphs. In addition, we also combined several AMR-to-text alignments with an attention mechanism and we supplemented the parser ... | ded41623ec63b47b4a90032b76c88300 | 2,019 | [
"our work involves enriching the stack - lstm transition - based amr parser ( ballesteros and al - onaizan , 2017 ) by augmenting training with policy learning and rewarding the smatch score of sampled graphs .",
"in addition , we also combined several amr - to - text alignments with an attention mechanism and we... | [
{
"event_type": "MDS",
"arguments": [
{
"text": "enriching",
"nugget_type": "E-PUR",
"argument_type": "Target",
"tokens": [
"enriching"
],
"offsets": [
3
]
},
{
"text": "training",
"nugget_type": "A... | [
"our",
"work",
"involves",
"enriching",
"the",
"stack",
"-",
"lstm",
"transition",
"-",
"based",
"amr",
"parser",
"(",
"ballesteros",
"and",
"al",
"-",
"onaizan",
",",
"2017",
")",
"by",
"augmenting",
"training",
"with",
"policy",
"learning",
"and",
"rewardi... |
ACL | Autoencoding Pixies: Amortised Variational Inference with Graph Convolutions for Functional Distributional Semantics | Functional Distributional Semantics provides a linguistically interpretable framework for distributional semantics, by representing the meaning of a word as a function (a binary classifier), instead of a vector. However, the large number of latent variables means that inference is computationally expensive, and trainin... | 68dd9d3c6696aa9d62b58cafbf37cd4b | 2,020 | [
"functional distributional semantics provides a linguistically interpretable framework for distributional semantics , by representing the meaning of a word as a function ( a binary classifier ) , instead of a vector .",
"however , the large number of latent variables means that inference is computationally expens... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "functional distributional semantics",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"functional",
"distributional",
"semantics"
],
"offsets": [
0,
... | [
"functional",
"distributional",
"semantics",
"provides",
"a",
"linguistically",
"interpretable",
"framework",
"for",
"distributional",
"semantics",
",",
"by",
"representing",
"the",
"meaning",
"of",
"a",
"word",
"as",
"a",
"function",
"(",
"a",
"binary",
"classifier... |
ACL | The Possible, the Plausible, and the Desirable: Event-Based Modality Detection for Language Processing | Modality is the linguistic ability to describe vents with added information such as how desirable, plausible, or feasible they are. Modality is important for many NLP downstream tasks such as the detection of hedging, uncertainty, speculation, and more. Previous studies that address modality detection in NLP often rest... | 13178d7ae0b12d85a0c61e77e7f454ea | 2,021 | [
"modality is the linguistic ability to describe vents with added information such as how desirable , plausible , or feasible they are .",
"modality is important for many nlp downstream tasks such as the detection of hedging , uncertainty , speculation , and more .",
"previous studies that address modality detec... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "modality",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"modality"
],
"offsets": [
0
]
}
],
"trigger": {
"text": "describe",
"tokens": ... | [
"modality",
"is",
"the",
"linguistic",
"ability",
"to",
"describe",
"vents",
"with",
"added",
"information",
"such",
"as",
"how",
"desirable",
",",
"plausible",
",",
"or",
"feasible",
"they",
"are",
".",
"modality",
"is",
"important",
"for",
"many",
"nlp",
"... |
ACL | Improving Arabic Diacritization with Regularized Decoding and Adversarial Training | Arabic diacritization is a fundamental task for Arabic language processing. Previous studies have demonstrated that automatically generated knowledge can be helpful to this task. However, these studies regard the auto-generated knowledge instances as gold references, which limits their effectiveness since such knowledg... | 86d45e0cba93e10df7621ec23220399d | 2,021 | [
"arabic diacritization is a fundamental task for arabic language processing .",
"previous studies have demonstrated that automatically generated knowledge can be helpful to this task .",
"however , these studies regard the auto - generated knowledge instances as gold references , which limits their effectivenes... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "arabic diacritization",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"arabic",
"diacritization"
],
"offsets": [
0,
1
]
}
],
"tr... | [
"arabic",
"diacritization",
"is",
"a",
"fundamental",
"task",
"for",
"arabic",
"language",
"processing",
".",
"previous",
"studies",
"have",
"demonstrated",
"that",
"automatically",
"generated",
"knowledge",
"can",
"be",
"helpful",
"to",
"this",
"task",
".",
"howe... |
ACL | Chase: A Large-Scale and Pragmatic Chinese Dataset for Cross-Database Context-Dependent Text-to-SQL | The cross-database context-dependent Text-to-SQL (XDTS) problem has attracted considerable attention in recent years due to its wide range of potential applications. However, we identify two biases in existing datasets for XDTS: (1) a high proportion of context-independent questions and (2) a high proportion of easy SQ... | 631b6e0352a0b6247c6f5cc6565589c7 | 2,021 | [
"the cross - database context - dependent text - to - sql ( xdts ) problem has attracted considerable attention in recent years due to its wide range of potential applications .",
"however , we identify two biases in existing datasets for xdts : ( 1 ) a high proportion of context - independent questions and ( 2 )... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "cross - database context - dependent text - to - sql ( xdts ) problem",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"cross",
"-",
"database",
"context",
"... | [
"the",
"cross",
"-",
"database",
"context",
"-",
"dependent",
"text",
"-",
"to",
"-",
"sql",
"(",
"xdts",
")",
"problem",
"has",
"attracted",
"considerable",
"attention",
"in",
"recent",
"years",
"due",
"to",
"its",
"wide",
"range",
"of",
"potential",
"app... |
ACL | Counterfactual Inference for Text Classification Debiasing | Today’s text classifiers inevitably suffer from unintended dataset biases, especially the document-level label bias and word-level keyword bias, which may hurt models’ generalization. Many previous studies employed data-level manipulations or model-level balancing mechanisms to recover unbiased distributions and thus p... | 5e45f8c571b949ca147e53a934d04e89 | 2,021 | [
"today ’ s text classifiers inevitably suffer from unintended dataset biases , especially the document - level label bias and word - level keyword bias , which may hurt models ’ generalization .",
"many previous studies employed data - level manipulations or model - level balancing mechanisms to recover unbiased ... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "text classifiers",
"nugget_type": "MOD",
"argument_type": "Target",
"tokens": [
"text",
"classifiers"
],
"offsets": [
3,
4
]
},
{
"text":... | [
"today",
"’",
"s",
"text",
"classifiers",
"inevitably",
"suffer",
"from",
"unintended",
"dataset",
"biases",
",",
"especially",
"the",
"document",
"-",
"level",
"label",
"bias",
"and",
"word",
"-",
"level",
"keyword",
"bias",
",",
"which",
"may",
"hurt",
"mo... |
ACL | FlipDA: Effective and Robust Data Augmentation for Few-Shot Learning | Most previous methods for text data augmentation are limited to simple tasks and weak baselines. We explore data augmentation on hard tasks (i.e., few-shot natural language understanding) and strong baselines (i.e., pretrained models with over one billion parameters). Under this setting, we reproduced a large number of... | 8d3807ad899ae6b80aa24c6d22c91af3 | 2,022 | [
"most previous methods for text data augmentation are limited to simple tasks and weak baselines .",
"we explore data augmentation on hard tasks ( i . e . , few - shot natural language understanding ) and strong baselines ( i . e . , pretrained models with over one billion parameters ) .",
"under this setting ,... | [
{
"event_type": "RWF",
"arguments": [
{
"text": "limited",
"nugget_type": "WEA",
"argument_type": "Fault",
"tokens": [
"limited"
],
"offsets": [
8
]
},
{
"text": "simple tasks",
"nugget_type": "WEA"... | [
"most",
"previous",
"methods",
"for",
"text",
"data",
"augmentation",
"are",
"limited",
"to",
"simple",
"tasks",
"and",
"weak",
"baselines",
".",
"we",
"explore",
"data",
"augmentation",
"on",
"hard",
"tasks",
"(",
"i",
".",
"e",
".",
",",
"few",
"-",
"s... |
ACL | A Good Prompt Is Worth Millions of Parameters: Low-resource Prompt-based Learning for Vision-Language Models | Large pre-trained vision-language (VL) models can learn a new task with a handful of examples and generalize to a new task without fine-tuning.However, these VL models are hard to deploy for real-world applications due to their impractically huge sizes and slow inference speed.To solve this limitation, we study prompt-... | 0b234c1228b475cea91b12a78f01d399 | 2,022 | [
"large pre - trained vision - language ( vl ) models can learn a new task with a handful of examples and generalize to a new task without fine - tuning .",
"however , these vl models are hard to deploy for real - world applications due to their impractically huge sizes and slow inference speed .",
"to solve thi... | [
{
"event_type": "RWF",
"arguments": [
{
"text": "vision - language models",
"nugget_type": "APP",
"argument_type": "Concern",
"tokens": [
"vision",
"-",
"language",
"models"
],
"offsets": [
4,
5... | [
"large",
"pre",
"-",
"trained",
"vision",
"-",
"language",
"(",
"vl",
")",
"models",
"can",
"learn",
"a",
"new",
"task",
"with",
"a",
"handful",
"of",
"examples",
"and",
"generalize",
"to",
"a",
"new",
"task",
"without",
"fine",
"-",
"tuning",
".",
"ho... |
ACL | Learning Disentangled Representations of Negation and Uncertainty | Negation and uncertainty modeling are long-standing tasks in natural language processing. Linguistic theory postulates that expressions of negation and uncertainty are semantically independent from each other and the content they modify. However, previous works on representation learning do not explicitly model this in... | e228d457b1cfdfe89896b5af11064430 | 2,022 | [
"negation and uncertainty modeling are long - standing tasks in natural language processing .",
"linguistic theory postulates that expressions of negation and uncertainty are semantically independent from each other and the content they modify .",
"however , previous works on representation learning do not expl... | [
{
"event_type": "MDS",
"arguments": [
{
"text": "variational autoencoder",
"nugget_type": "MOD",
"argument_type": "TriedComponent",
"tokens": [
"variational",
"autoencoder"
],
"offsets": [
65,
66
]
},... | [
"negation",
"and",
"uncertainty",
"modeling",
"are",
"long",
"-",
"standing",
"tasks",
"in",
"natural",
"language",
"processing",
".",
"linguistic",
"theory",
"postulates",
"that",
"expressions",
"of",
"negation",
"and",
"uncertainty",
"are",
"semantically",
"indepe... |
ACL | ePiC: Employing Proverbs in Context as a Benchmark for Abstract Language Understanding | While large language models have shown exciting progress on several NLP benchmarks, evaluating their ability for complex analogical reasoning remains under-explored. Here, we introduce a high-quality crowdsourced dataset of narratives for employing proverbs in context as a benchmark for abstract language understanding.... | f5855edc34938bda6d0e9b9622249f5a | 2,022 | [
"while large language models have shown exciting progress on several nlp benchmarks , evaluating their ability for complex analogical reasoning remains under - explored .",
"here , we introduce a high - quality crowdsourced dataset of narratives for employing proverbs in context as a benchmark for abstract langua... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
27
]
},
{
"text": "high - quality crowdsourced dataset of narratives"... | [
"while",
"large",
"language",
"models",
"have",
"shown",
"exciting",
"progress",
"on",
"several",
"nlp",
"benchmarks",
",",
"evaluating",
"their",
"ability",
"for",
"complex",
"analogical",
"reasoning",
"remains",
"under",
"-",
"explored",
".",
"here",
",",
"we"... |
ACL | Contrastive Visual Semantic Pretraining Magnifies the Semantics of Natural Language Representations | We examine the effects of contrastive visual semantic pretraining by comparing the geometry and semantic properties of contextualized English language representations formed by GPT-2 and CLIP, a zero-shot multimodal image classifier which adapts the GPT-2 architecture to encode image captions. We find that contrastive ... | b1b52caa59e9895d6f3ea81fb3b9f1ba | 2,022 | [
"we examine the effects of contrastive visual semantic pretraining by comparing the geometry and semantic properties of contextualized english language representations formed by gpt - 2 and clip , a zero - shot multimodal image classifier which adapts the gpt - 2 architecture to encode image captions .",
"we find... | [
{
"event_type": "PUR",
"arguments": [
{
"text": "effects of contrastive visual semantic pretraining",
"nugget_type": "TAK",
"argument_type": "Aim",
"tokens": [
"effects",
"of",
"contrastive",
"visual",
"semantic",
... | [
"we",
"examine",
"the",
"effects",
"of",
"contrastive",
"visual",
"semantic",
"pretraining",
"by",
"comparing",
"the",
"geometry",
"and",
"semantic",
"properties",
"of",
"contextualized",
"english",
"language",
"representations",
"formed",
"by",
"gpt",
"-",
"2",
"... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.