Papers

Filtered to PPO · clear filter

Browse by term

continual learning 83reinforcement learning 50large language models 12benchmarking 11benchmarks 11language models 11vision-language models 11robotics 7natural language processing 6world models 6generative models 5recursive self-improvement 5attention mechanisms 4diffusion Transformers 4multi-agent systems 4multimodal learning 4multimodal models 4on-policy distillation 4self-distillation 4self-supervised learning 4transformers 4video generation 4vision-language-action models 4agent-based systems 3agentic models 3agentic search 3autonomous systems 3coding agents 3diffusion models 3image generation 3

Matching papers

Rethinking Critic Learning in PPO: Understanding and Mitigating Value Flattening

60 upvotes · 16 SEP 2026 · Yizhuo Li, Jianhao Yan, Yun Luo et al.

This paper investigates a common problem in reinforcement learning for language models called Value Flattening, where critics fail to accurately estimate state values, and proposes a new method, SP^3O, to mitigate this issue by supervising only a few well-separated states per response.

reinforcement learningPPOvalue estimationstate spacesparse supervisioncritic learning