CARM: Cancellation-Aware Response Masking for LLM Reinforcement Learning
Recent years have witnessed the rapid adoption of reinforcement learning (RL) in large language model (LLM) post-training, with substantial gains in mathematical reasoning and code generation. In practical systems, however, policy updates and differences between rollout and training engines can make sampled responses o...
Ya-Fei Zhang, Song-Shuo Lu, Si-Cong Liao et al.
· 0 citations