2 papers
cs.AI2025
StepHint: Multi-level Stepwise Hints Enhance Reinforcement Learning to Reason
Kaiyi Zhang, Ang Lv, Jinpeng Li +4
Reinforcement learning with verifiable rewards (RLVR) is a promising approach for improving the complex reasoning abilities of large language models (LLMs). However, current RLVR m…
cs.CL2024
PEAR: Position-Embedding-Agnostic Attention Re-weighting Enhances Retrieval-Augmented Generation with Zero Inference Overhead
Tao Tan, Yining Qian, Ang Lv +7
Large language models (LLMs) enhanced with retrieval-augmented generation (RAG) have introduced a new paradigm for web search. However, the limited context awareness of LLMs degrad…