9 papers
What Does Multi-Turn Reinforcement Learning Really Improve in LLM Agents?
Kejian Zhu, Zhuoran Jin, Shangqing Tu +5
While multi-turn agentic reinforcement learning (RL) excels on long-horizon tasks, its performance gains remain highly sensitive to interaction constraints. Evaluating across sampl…
SFT Conflicts, RL Coexists: A Theoretical and Empirical Analysis of Multi-Task Learning for LLMs
Kejian Zhu, Zhuoran Jin, Shangqing Tu +5
Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL) exhibit fundamentally different behaviors in enhancing multi-task reasoning for large language models (LLMs). Our preli…
Beyond Simply Environment Scaling: Designing Effective Environment Distributions for Multimodal Agent Learning
Kejian Zhu, Zhuoran Jin, Dongqi Huang +4
Recent works train agents by constructing large-scale multimodal environment pools. However, we find that simply increasing the number of multimodal environments does not always be…
Look Light, Think Heavy: What Multimodal Chain-of-Thought Reasoning Can and Cannot Do
Zhuoran Jin, Kejian Zhu, Hongbang Yuan +5
Chain-of-Thought (CoT) has become a standard method for improving reasoning capabilities in large language models (LLMs) by eliciting step-by-step thinking, but its effectiveness i…
Agentic Environment Engineering for Large Language Models: A Survey of Environment Modeling, Synthesis, Evaluation, and Application
Jiachun Li, Zhuoran Jin, Tianyi Men +12
Environments serve as interactive systems for large language model (LLM) based agents across diverse scenarios and play a crucial role in driving the continual evolution of model c…
Omni-Reward: Towards Generalist Omni-Modal Reward Modeling with Free-Form Preferences
Zhuoran Jin, Hongbang Yuan, Kejian Zhu +5
Reward models (RMs) play a critical role in aligning AI behaviors with human preferences, yet they face two fundamental challenges: (1) Modality Imbalance, where most RMs are mainl…
Establishing Trustworthy LLM Evaluation via Shortcut Neuron Analysis
Kejian Zhu, Shangqing Tu, Zhuoran Jin +3
The development of large language models (LLMs) depends on trustworthy evaluation. However, most current evaluations rely on public benchmarks, which are prone to data contaminatio…
MMR-V: What's Left Unsaid? A Benchmark for Multimodal Deep Reasoning in Videos
Kejian Zhu, Zhuoran Jin, Hongbang Yuan +6
The sequential structure of videos poses a challenge to the ability of multimodal large language models (MLLMs) to locate multi-frame evidence and conduct multimodal reasoning. How…
DICE: Detecting In-distribution Contamination in LLM's Fine-tuning Phase for Math Reasoning
Shangqing Tu, Kejian Zhu, Yushi Bai +3
The advancement of large language models (LLMs) relies on evaluation using public benchmarks, but data contamination can lead to overestimated performance. Previous researches focu…