This paper proposes a method to improve long-CoT reasoning in large language models by addressing the issue of unequal token contributions to the final outcome. It shows that current methods, such as GRPO, assign too much credit to highly sensitive tokens and proposes a new method, CSCR, that reduces credit for these tokens to improve performance.
Firehose
Filtered to tagged “verifiable rewards” · clear filters
Browse: People · Companies · Papers · Podcasts · Hacker News · Deep dives
Browse by tag
artificial intelligence 57continual learning 24agentic coding 21open-weight models 20AI 16AI agents 13reinforcement learning 11cybersecurity 9AI safety 8finance 8language models 7large language models 7open-source 7productivity 7deep learning 6machine learning 6natural language processing 6Reinforcement learning 6tech 6Databricks 5robotics 5software development 5Agentic AI 4benchmarking 4Diffusion models 4multi-agent systems 4Recursive self-improvement 4world models 4AI ethics 3AI infrastructure 3