5 papers
Evolving from Tool User to Creator via Training-Free Experience Reuse in Multimodal Reasoning
Xintian Shen, Jiawei Chen, Lihao Zheng +3
Existing Tool-Integrated Reasoning (TIR) models have effectively extended the question-answering capabilities of LLMs by incorporating external tools. However, real-world scenarios…
MIRG-RL: Multi-Image Reasoning and Grounding with Reinforcement Learning
Lihao Zheng, Jiawei Chen, Xintian Shen +2
Multi-image reasoning and grounding require understanding complex cross-image relationships at both object levels and image levels. Current Large Visual Language Models (LVLMs) fac…
COPO: Consistency-Aware Policy Optimization
Jinghang Han, Jiawei Chen, Hang Shao +7
Reinforcement learning has significantly enhanced the reasoning capabilities of Large Language Models (LLMs) in complex problem-solving tasks. Recently, the introduction of DeepSee…
WereWolf-Plus: An Update of Werewolf Game setting Based on DSGBench
Xinyuan Xia, Yuanyi Song, Haomin Ma +1
With the rapid development of LLM-based agents, increasing attention has been given to their social interaction and strategic reasoning capabilities. However, existing Werewolf-bas…
Pseudo-Knowledge Graph: Meta-Path Guided Retrieval and In-Graph Text for RAG-Equipped LLM
Yuxin Yang, Haoyang Wu, Tao Wang +3
The advent of Large Language Models (LLMs) has revolutionized natural language processing. However, these models face challenges in retrieving precise information from vast dataset…