4 papers
The Easy, the Hard, and the Learnable: Confidence and Difficulty-Adaptive Policy Optimization for LLM Reasoning
Zhanke Zhou, Xiangyu Lu, Chentao Cao +4
RL with verifiable rewards can substantially improve LLM reasoning, yet standard GRPO-style training often treats easy, hard, and learnable questions alike through uniform sampling…
Learning Locally, Revising Globally: Global Reviser for Federated Learning with Noisy Labels
Yuxin Tian, Mouxing Yang, Yuhao Zhou +5
Conventional federated learning (FL) heavily depends on high-quality labels, which are often impractical in the real world, leading to the federated label-noise (F-LN) problem. Wor…
Reinforcement Learning with Verifiable yet Noisy Rewards under Imperfect Verifiers
Xin-Qiang Cai, Wei Wang, Feng Liu +3
Reinforcement Learning with Verifiable Rewards (RLVR) replaces costly human labeling with automated verifiers. To reduce verifier hacking, many RLVR systems binarize rewards to $\{…
Combating Noisy Labels through Fostering Self- and Neighbor-Consistency
Zeren Sun, Yazhou Yao, Tongliang Liu +3
Label noise is pervasive in various real-world scenarios, posing challenges in supervised deep learning. Deep networks are vulnerable to such label-corrupted samples due to the mem…