{"arxiv_id":"2602.14697","title":"Evolutionary System Prompt Learning for Reinforcement Learning in LLMs","abstract":"Building agentic systems that can autonomously self-improve from experience is a longstanding goal of AI. Large language models (LLMs) today primarily self-improve via two mechanisms: self-reflection for context updates, and reinforcement learning (RL) for weight updates. In this work, we propose Evolutionary System Prompt Learning (E-SPL), a method for jointly improving a model's experiential context and its weights. In each RL iteration, E-SPL samples trajectories under multiple system prompts in parallel, then uses them to jointly update the weights with RL and evolve the system prompts with self-reflection. E-SPL encourages a natural division between procedural knowledge in weights and interpretable declarative knowledge in system prompts. Most notably, in an easy-to-hard generalization setting (AIME $\\rightarrow$ BeyondAIME), E-SPL improves RL success rate from 38.8% $\\rightarrow$ 45.1%. E-SPL also improves RL on AIME 2025 (56.3% $\\rightarrow$ 60.6%), HMMT 2025 (50.0% $\\rightarrow$ 52.7%), and agentic search (44.2% $\\rightarrow$ 48.6% on gpt-oss-120b). Across all settings, E-SPL outperforms both RL-only and evolution-only baselines, demonstrating that weight updates and context updates are deeply synergistic and can together yield gains in generalization that neither achieves alone. Code is available at https://github.com/LunjunZhang/E-SPL.","primary_cat":"cs.AI","cross_cats":["cs.LG"],"submitted_at":"2026-02-16T12:34:27+00:00","latest_version":5,"source":"arxiv","source_label":"arXiv","source_url":null,"has_pdf":false,"has_source":false,"verdict":null,"confidence":null,"one_line_summary":null,"novelty_score":null,"correctness_risk":null,"formal_verification":null,"upvotes":0,"downvotes":0,"pith_score":0,"inbound_count":4,"_links":{"html":"https://pith.science/paper/2602.14697","verdict":"/api/v1/verdicts/2602.14697"}}