Soujanya Poria's picture

Soujanya Poria

soujanyaporia

·

https://soujanyaporia.github.io/

AI & ML interests

Natural Language Processing

Organizations

commented a paper 3 months ago

Error-Free Linear Attention is a Free Lunch: Exact Solution from Continuous-Time Dynamics

Paper • 2512.12602 • Published Dec 14, 2025 • 44 •

New activity in declare-lab/nora-1.5 5 months ago

Add GitHub code link to model card

#1 opened 5 months ago by

commented 2 papers 5 months ago

NORA-1.5: A Vision-Language-Action Model Trained using World Model- and Action-based Preference Rewards

Paper • 2511.14659 • Published Nov 18, 2025 • 13 •

10 Open Challenges Steering the Future of Vision-Language-Action Models

Paper • 2511.05936 • Published Nov 8, 2025 • 6 •

New activity in declare-lab/OffTopicEval 6 months ago

[bot] Conversion to Parquet

#1 opened 7 months ago by

parquet-converter

commented a paper 6 months ago

Demystifying deep search: a holistic evaluation with hint-free multi-hop questions and factorised metrics

Paper • 2510.05137 • Published Oct 1, 2025 • 6 •

New activity in declare-lab/OffTopicEval 7 months ago

Improve dataset card: Add metadata (task categories, language, tags) and links to paper & code

#2 opened 7 months ago by

commented 2 papers 7 months ago

Training Vision-Language Process Reward Models for Test-Time Scaling in Multimodal Reasoning: Key Insights and Lessons Learned

Paper • 2509.23250 • Published Sep 27, 2025 • 6 •

OffTopicEval: When Large Language Models Enter the Wrong Chat, Almost Always!

Paper • 2509.26495 • Published Sep 30, 2025 • 13 •

New activity in declare-lab/KAIROS_EVAL 8 months ago

Improve KAIROS_EVAL dataset card: Add paper/code links, tags, and update sample usage

#2 opened 8 months ago by

commented a paper 9 months ago

JAM: A Tiny Flow-based Song Generator with Fine-grained Controllability and Aesthetic Alignment

Paper • 2507.20880 • Published Jul 28, 2025 • 11 •

New activity in declare-lab/Trust-Data 11 months ago

Improve dataset card

#2 opened about 1 year ago by

commented a paper 11 months ago

Error Typing for Smarter Rewards: Improving Process Reward Models with Error-Aware Hierarchical Supervision

Paper • 2505.19706 • Published May 26, 2025 • 3 •

commented a paper 12 months ago

NORA: A Small Open-Sourced Generalist Vision Language Action Model for Embodied Tasks

Paper • 2504.19854 • Published Apr 28, 2025 • 7 •

New activity in declare-lab/AlgoPuzzleVQA about 1 year ago

Librarian Bot: Add language metadata for dataset

#2 opened about 1 year ago by

New activity in declare-lab/TangoFlux over 1 year ago

Two fighter jets are fighting with missile. missile launching followed by an explosion

#1 opened about 2 years ago by

a disciplined military marching

#5 opened almost 2 years ago by

commented 3 papers over 1 year ago

TangoFlux: Super Fast and Faithful Text to Audio Generation with Flow Matching and Clap-Ranked Preference Optimization

Paper • 2412.21037 • Published Dec 30, 2024 • 24 •

Measuring and Enhancing Trustworthiness of LLMs in RAG through Grounded Attributions and Learning to Refuse

Paper • 2409.11242 • Published Sep 17, 2024 • 7 •

Ferret: Faster and Effective Automated Red Teaming with Reward-Based Scoring Technique

Paper • 2408.10701 • Published Aug 20, 2024 • 12 •