5 papers
Critique to Verify: Accurate and Honest Test-Time Scaling with RL-Trained Verifiers
Zhicheng Yang, Zhijiang Guo, Yinya Huang +4
Test-time scaling via solution sampling and aggregation has become a key paradigm for improving the reasoning performance of Large Language Models (LLMs). While reward model select…
When Inverse Data Outperforms: Exploring the Pitfalls of Mixed Data in Multi-Stage Fine-Tuning
Mengyi Deng, Xin Li, Tingyu Zhu +3
Existing work has shown that o1-level performance can be achieved with limited data distillation, but most existing methods focus on unidirectional supervised fine-tuning (SFT), ov…
Understanding GUI Agent Localization Biases through Logit Sharpness
Xingjian Tao, Yiwei Wang, Yujun Cai +2
Multimodal large language models (MLLMs) have enabled GUI agents to interact with operating systems by grounding language into spatial actions. Despite their promising performance,…
TreeRPO: Tree Relative Policy Optimization
Zhicheng Yang, Zhijiang Guo, Yinya Huang +3
Large Language Models (LLMs) have shown remarkable reasoning capabilities through Reinforcement Learning with Verifiable Rewards (RLVR) methods. However, a key limitation of existi…
Are LLMs Really Not Knowledgeable? Mining the Submerged Knowledge in LLMs' Memory
Xingjian Tao, Yiwei Wang, Yujun Cai +2
Large language models (LLMs) have shown promise as parametric knowledge bases, but often underperform on question answering (QA) tasks due to hallucinations and uncertainty. While…