让提示词优化可追踪、可定位、可行动，通过结构化编辑提升效果。
SEPO: Evidence-Grounded Prompt Optimization via Structural Editing
- 按结构单元局部修改提示词，记录每步修改的因果影响。
- 在14个任务上超越最强基线3.1个百分点，准确率分别达61.9%和73.3%。
- 优化耗能更低，生成提示短5倍，适合追求高效可控的开发者。
现有仅通过API的提示词优化方法常被称作可解释，但实际仅支持事后检查：每次迭代仍以整体字符串形式重写提示词，仅留下完整提示的差异痕迹，无法定位具体修改。本文提出SEPO（结构化、证据驱动的提示词优化），一种基于编辑效应谱系反馈的多轨迹优化框架。它不将每次迭代视为孤立的全提示重写，而是对双层提示结构中的稳定类型单元进行局部编辑，将每项编辑的目标与实际效果关联到其新修复或破坏的示例，并将此编辑-效果记录传递至同搜索分支的后续架构调用中。这使提示优化具备可追溯性、可归因性和可操作性。在14个独立任务测试集上，SEPO相比最强基线GEPA，在Llama-3.1-8B-Instruct上提升3.1个百分点，在Qwen3-8B上提升2.2个百分点，达到61.9%和73.3%的宏平均准确率。同时，SEPO位于优化时间和测试时间的帕累托前沿，仅消耗290万优化标记（vs GEPA的410万），生成提示长度超过5倍缩短。
原文摘要 · Abstract (English)
Existing API-only prompt optimisers are often described as interpretable, but in practice, this usually means only post-hoc inspectability: each iteration still rewrites the prompt as one opaque string, leaving a trace of full-prompt diffs rather than localisable, machine-readable edits. This paper introduces SEPO (Structural, Evidence-grounded Prompt Optimization), a multi-trajectory prompt optimiser centred on edit-effect lineage feedback. Rather than treating each iteration as an isolated whole-prompt rewrite, SEPO locally edits stable, typed units in a two-layer prompt schema, links the target and realised structural operations of each edit to the examples it newly fixes or breaks, and carries this edit-effect record forward to guide later architect calls on the same search branch. This makes prompt optimisation addressable, attributable, and actionable. Across a 14-task held-out suite, SEPO improves over the strongest baseline, GEPA, by 3.1 pp on Llama-3.1-8B-Instruct and 2.2 pp on Qwen3-8B, reaching 61.9% and 73.3% macro accuracy. SEPO also lies on both the optimisation-time and test-time Pareto frontiers, spending 2.9M optimisation tokens versus 4.1M for GEPA and producing prompts over 5x shorter.
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。