E arl : Efficient Agentic RL Post-Training for LLMs under Dynamic Context Lengths

Agentic reinforcement learning (RL) is increasingly used to post-train large language models (LLMs) into multi-turn agents. However, scaling agentic RL is challenging because the effective context length grows during training: agents become more verbose and rollouts accumulate long histories. This context-length explosion creates two coupled systems bottlenecks: (i) attention compute and memory scale with length, causing truncation or out-of-memory (OOM) events that can destabilize learning; and (ii) experience tensors (e.g., tokens and log-probabilities) scale with length and make cross-device exchange a dominant cost.