mlsanityData to 7 Oct 2026

SEPO: Evidence-Grounded Prompt Optimization via Structural Editing

arXiv 2608.28067 ↗First posted 28 Aug 2026cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 29 of its 52 references that are in the map. First posted on arXiv 28 Aug 2026.

Builds on

29 of 52 references.

References · 29

  1. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  2. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,686
  3. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,317
  4. 2022Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Themcited by 1,767
  5. 2024MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmarkcited by 1,758
  6. 2023Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generationcited by 1,332
  7. 2022Large Language Models Are Human-Level Prompt Engineerscited by 1,020
  8. 2023Large Language Models as Optimizerscited by 830
  9. 2023DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelinescited by 803
  10. 2023Automatic Prompt Optimization with "Gradient Descent" and Beam Searchcited by 628
  11. 2023Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolutioncited by 521
  12. 2024Optimizing Instructions and Demonstrations for Multi-Stage Language Model Programscited by 270
17 more
  1. 2023PromptAgent: Strategic Planning with Language Models Enables Expert-level Prompt Optimizationcited by 226
  2. 2025AgenTracer: Who Is Inducing Failure in the LLM Agentic Systems?cited by 105
  3. 2025A Systematic Survey of Automatic Prompt Optimization Techniquescited by 61
  4. 2024PRompt Optimization in Multi-Step Tasks (PROMST): Integrating Human Feedback and Heuristic-based Samplingcited by 32
  5. 2024Symbolic Prompt Program Search: A Structure-Aware Approach to Efficient Compile-Time Prompt Optimizationcited by 29
  6. 2024CriSPO: Multi-Aspect Critique-Suggestion-guided Automatic Prompt Optimization for Text Generationcited by 18
  7. 2024Task Facet Learning: A Structured Approach to Prompt Optimizationcited by 17
  8. 2024SCULPT: Systematic Tuning of Long Promptscited by 8
  9. 2025ZERA: Zero-init Instruction Evolving Refinement Agent -- From Zero Instructions to Structured Prompts via Principle-based Optimizationcited by 6
  10. 2026Modular Prompt Optimization: Optimizing Structured Prompts with Section-Local Textual Gradientscited by 5
  11. 2026Adaptive Prompt Structure Factorization: A Framework for Self-Discovering and Optimizing Compositional Prompt Programscited by 5
  12. 2025Local Prompt Optimizationcited by 4
  13. 2026Learning from Prompt itself: the Hierarchical Attribution Prompt Optimizationcited by 1
  14. 2026Select Smarter, Not More: Prompt-Aware Evaluation Scheduling with Submodular Guaranteescited by 1
  15. 2026One Rewrite to Fix Them All? Type-Aware Repair Allocation for Text-to-Image Prompt Optimizationcited by 1
  16. 2026Are Prompt Optimizers Blind? Cross-Modal Visual Feedback for Automatic Prompt Optimizationcited by 1
  17. 2026Plans You Can Check: Verifier-Grounded Learning of an Open-Weight Planner for Executable Video-Editingcited by 1

Built on by

Not cited yet.