让提示代理自我进化，自动优化自身和任务的提示词。
SePO: Self-Evolving Prompt Agent for System Prompt Optimization
- 提示代理同时优化自身与任务的系统提示，采用自指设计。
- 在5个基准上平均准确率比人工思维链高4.49分。
- 预训练技能可泛化到未见过的任务，不依赖记忆提示。
系统提示优化可在不修改模型的前提下改善智能体行为，生成人类可读、模型无关的指令。现有方法虽构建提示代理来优化任务代理的系统提示，但提示代理自身的系统提示仍需手工设计且固定不变。本文提出自演化提示优化（SePO），将提示代理自身的系统提示也作为优化目标。SePO采用自指设计：单一提示代理在开放式演化搜索中同时优化任务代理和自身的系统提示，维护候选提示档案作为演进跳板。训练分为两个阶段：预训练在多任务数据池上演化提示代理，微调则应用于目标任务。在涵盖数学（AIME'25）、抽象推理（ARC-AGI-1）、研究生级科学（GPQA）、代码生成（MBPP）和逻辑谜题（Sudoku）的五个基准上，SePO持续优于Manual-CoT、TextGrad和MetaSPO，平均准确率相比Manual-CoT提升4.49分。预训练获得的提示优化能力可泛化至预训练混合任务之外，而非记忆特定任务提示。
原文摘要 · Abstract (English)
System prompt optimization improves agent behavior without modifying the underlying model, yielding human-readable, model-agnostic instructions. Existing methods build a prompt agent that refines task agents' system prompts, yet leave the prompt agent's own system prompt hand-engineered and fixed. We propose Self-Evolving Prompt Optimization (SePO), which treats the prompt agent's own system prompt as an optimization target alongside task agents' system prompts. SePO adopts a self-referential design. A single prompt agent improves both task agents' system prompts and its own under an open-ended evolutionary search that maintains an archive of candidate prompts as stepping stones. Training proceeds in two stages: pre-training evolves the prompt agent on a multi-task pool, and fine-tuning then applies it to a target task. Across five benchmarks spanning math (AIME'25), abstract reasoning (ARC-AGI-1), graduate-level science (GPQA), code generation (MBPP), and logic puzzles (Sudoku), SePO consistently outperforms Manual-CoT, TextGrad, and MetaSPO, improving the average accuracy by 4.49 points compared to Manual-CoT. The prompt optimization skill from pre-training also generalizes to tasks beyond the pre-training mixture, rather than memorizing per-task prompts.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。