{"ID":2859358,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.05943","arxiv_id":"2510.05943","title":"EARL: Efficient Agentic Reinforcement Learning Systems for Large Language Models","abstract":"Reinforcement learning (RL) has become a pivotal component of large language model (LLM) post-training, and agentic RL extends this paradigm to operate as agents through multi-turn interaction and tool use. Scaling such systems exposes two practical bottlenecks: (1) context length grows rapidly during training, inflating memory usage and latency, and triggering out-of-memory (OOM) failures; and (2) intermediate tensors accumulate with context length, making cross-device data movement a major system bottleneck. We present EARL, a scalable system for efficient agentic RL. EARL designs a parallelism selector that dynamically adapts model and training parallelism across RL stages based on sequence length and system load, and a data dispatcher that performs layout-aware, decentralized exchange of intermediate data batches. Together, these components increase throughput, reduce long-context failures, and enable stable large-scale training of agentic LLMs without relying on hard limits or penalties of context length.","short_abstract":"Reinforcement learning (RL) has become a pivotal component of large language model (LLM) post-training, and agentic RL extends this paradigm to operate as agents through multi-turn interaction and tool use. Scaling such systems exposes two practical bottlenecks: (1) context length grows rapidly during training, inflati...","url_abs":"https://arxiv.org/abs/2510.05943","url_pdf":"https://arxiv.org/pdf/2510.05943v1","authors":"[\"Zheyue Tan\",\"Mustapha Abdullahi\",\"Tuo Shi\",\"Huining Yuan\",\"Zelai Xu\",\"Chao Yu\",\"Boxun Li\",\"Bo Zhao\"]","published":"2025-10-07T13:52:51Z","proceeding":"cs.DC","tasks":"[\"cs.DC\",\"cs.LG\"]","methods":"[\"Reinforcement Learning\",\"Large Language Model\",\"Language Model\"]","has_code":false}