Deep Cognition and Language Research (DeCLaRe) Lab

university

https://declare-lab.github.io

AI & ML interests

Natural Language Processing

Recent Activity

soujanyaporia updated a dataset 1 day ago

declare-lab/rq-bench

soujanyaporia submitted a paper 1 day ago

On the Limits of LLM-as-Judge for Scientific Novelty Assessment

Tej3 authored a paper 4 days ago

GRAIL: Gradient-Reweighted Advantages for Reinforcement Learning with Verifiable Rewards

View all activity

Papers

On the Limits of LLM-as-Judge for Scientific Novelty Assessment

GRAIL: Gradient-Reweighted Advantages for Reinforcement Learning with Verifiable Rewards

View all Papers

declare-lab 's papers 8

Submitted by

Soujanya Poria

On the Limits of LLM-as-Judge for Scientific Novelty Assessment

declare-lab

Deep Cognition and Language Research (DeCLaRe) Lab

Submitted by

Soujanya Poria

GRAIL: Gradient-Reweighted Advantages for Reinforcement Learning with Verifiable Rewards

declare-lab

Deep Cognition and Language Research (DeCLaRe) Lab

Submitted by

Soujanya Poria

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

declare-lab

Deep Cognition and Language Research (DeCLaRe) Lab

Submitted by

Soujanya Poria

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

declare-lab

Deep Cognition and Language Research (DeCLaRe) Lab

Submitted by

Soujanya Poria

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

declare-lab

Deep Cognition and Language Research (DeCLaRe) Lab

Submitted by

Soujanya Poria

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

declare-lab

Deep Cognition and Language Research (DeCLaRe) Lab

Submitted by

Soujanya Poria

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

declare-lab

Deep Cognition and Language Research (DeCLaRe) Lab

Submitted by

Soujanya Poria

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

declare-lab

Deep Cognition and Language Research (DeCLaRe) Lab