Papers

Filtered to agentic reinforcement learning · clear filter

Browse by term

continual learning 82reinforcement learning 50large language models 12benchmarking 11benchmarks 11language models 11vision-language models 10robotics 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

AgentOPSD: Recursive Self-Distillation for Agentic Reinforcement Learning

85 upvotes · 6 AUG 2026 · Zi-Han Wang, Zhengxi Lu, Zhiyuan Yao et al.

This paper proposes a method called AgentOPSD for turn-level credit assignment in agentic reinforcement learning, which helps to identify pivotal decisions that determine outcomes in long-horizon tasks. Practitioners may care about this because it can improve the performance of reinforcement learning agents in complex tasks.

agentic reinforcement learningturn-level credit assignmentrecursive self-distillationcritic-free methodsBayesian belief states