mlsanityData to 14 Oct 2026

Self-Supervised Prompt Optimization

arXiv 2502.06855 ↗First posted 7 Feb 2025cs.CL cs.AI cs.LGcited by 37

On the map: LLM reasoning → Prompting and chain-of-thought reasoning

Abstract (arXiv)

Well-designed prompts are crucial for enhancing Large language models' (LLMs) reasoning capabilities while aligning their outputs with task requirements across diverse domains. However, manually designed prompts require expertise and iterative experimentation. While existing prompt optimization methods aim to automate this process, they rely heavily on external references such as ground truth or by humans, limiting their applicability in real-world scenarios where such data is unavailable or costly to obtain. To address this, we propose Self-Supervised Prompt Optimization (SPO), a cost-efficient framework that discovers effective prompts for both closed and open-ended tasks without requiring external reference. Motivated by the observations that prompt quality manifests directly in LLM outputs and LLMs can effectively assess adherence to task requirements, we derive evaluation and optimization signals purely from output comparisons. Specifically, SPO selects superior prompts through pairwise output comparisons evaluated by an LLM evaluator, followed by an LLM optimizer that aligns outputs with task requirements. Extensive experiments demonstrate that SPO outperforms state-of-the-art prompt optimization methods, achieving comparable or superior results with significantly lower costs (e.g., 1.1% to 5.6% of existing methods) and fewer samples (e.g., three samples). The code is available at https://github.com/FoundationAgents/SPO.

Key facts. Cited by 37 papers in the map (data through 14 Oct 2026). Placed in Prompting and chain-of-thought reasoning, one of 17 directions in LLM reasoning. Builds on 47 of its 56 references that are in the map. First posted on arXiv 7 Feb 2025.

Builds on

47 of 56 references.

Closest ancestors · 5

Same direction, newest first

  1. 2023Take a Step Back: Evoking Reasoning via Abstraction in Large Language Models9 Oct
  2. 2023Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution28 Sep
  3. 2022Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them17 Oct
  4. 2022Large Language Models are Zero-Shot Reasoners24 May
  5. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Models28 Jan

Other references · 42

  1. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  2. 2023GPQA: A Graduate-Level Google-Proof Q&A Benchmarkcited by 2,879
  3. 2023Voyager: An Open-Ended Embodied Agent with Large Language Modelscited by 2,191
  4. 2023MetaGPT: Meta Programming for A Multi-Agent Collaborative Frameworkcited by 1,808
  5. 2022Large Language Models Are Human-Level Prompt Engineerscited by 1,045
  6. 2023Large Language Models as Optimizerscited by 864
  7. 2023DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelinescited by 832
  8. 2023Large Language Modelscited by 780
  9. 2024Large Language Model based Multi-Agents: A Survey of Progress and Challengescited by 769
  10. 2023AGIEval: A Human-Centric Benchmark for Evaluating Foundation Modelscited by 716
  11. 2023Automatic Prompt Optimization with "Gradient Descent" and Beam Searchcited by 653
  12. 2024From Generation to Judgment: Opportunities and Challenges of LLM-as-a-judgecited by 494
30 more
  1. 2025Agent Laboratory: Using LLM Agents as Research Assistantscited by 425
  2. 2017"Liar, Liar Pants on Fire": A New Benchmark Dataset for Fake News Detectioncited by 363
  3. 2023PromptAgent: Strategic Planning with Language Models Enables Expert-level Prompt Optimizationcited by 233
  4. 2024Data Interpreter: An LLM Agent For Data Sciencecited by 225
  5. 2023EvoPrompt: Connecting LLMs with Evolutionary Algorithms Yields Powerful Prompt Optimizerscited by 193
  6. 2024Aligning with Human Judgement: The Role of Pairwise Preference in Large Language Model Evaluatorscited by 149
  7. 2024Does Prompt Formatting Have Any Impact on LLM Performance?cited by 141
  8. 2023Rephrase and Respond: Let Large Language Models Ask Better Questions for Themselvescited by 123
  9. 2024Code Generation with AlphaCodium: From Prompt Engineering to Flow Engineeringcited by 103
  10. 2024The Dawn of Natural Language to SQL: Are We Fully Ready?cited by 79
  11. 2024Archon: An Architecture Search Framework for Inference-Time Techniquescited by 60
  12. 2024A Survey of Text-to-SQL in the Era of LLMs: Where are we, and where are we going?cited by 56
  13. 2024ChartInsights: Evaluating Multimodal Large Language Models for Low-Level Chart Question Answeringcited by 50
  14. 2024Efficient Prompting Methods for Large Language Models: A Surveycited by 47
  15. 2025Test-Time Preference Optimization: On-the-Fly Alignment via Iterative Textual Feedbackcited by 41
  16. 2024CharacterBox: Evaluating the Role-Playing Capabilities of LLMs in Text-Based Virtual Worldscited by 38
  17. 2024Prompt Optimization with Human Feedbackcited by 35
  18. 2024PRompt Optimization in Multi-Step Tasks (PROMST): Integrating Human Feedback and Heuristic-based Samplingcited by 32
  19. 2024MobileExperts: A Dynamic Tool-Enabled Agent Team in Mobile Devicescited by 29
  20. 2024Are Large Language Models Good Statisticians?cited by 27
  21. 2024Fairer Preferences Elicit Improved Human-Aligned Large Language Model Judgmentscited by 25
  22. 2024Generative AI for Visualization: State of the Art and Future Directionscited by 23
  23. 2024NegativePrompt: Leveraging Psychology for Large Language Models Enhancement via Negative Emotional Stimulicited by 18
  24. 2024CriSPO: Multi-Aspect Critique-Suggestion-guided Automatic Prompt Optimization for Text Generationcited by 18
  25. 2024Efficient and Accurate Prompt Optimization: the Benefit of Memory in Exemplar-Guided Reflectioncited by 18
  26. 2024How to Correctly do Semantic Backpropagation on Language-based Agentic Systemscited by 18
  27. 2024StraGo: Harnessing Strategic Guidance for Prompt Optimizationcited by 17
  28. 2024AskChart: Universal Chart Understanding through Textual Enhancementcited by 12
  29. 2024Data Playwright: Authoring Data Videos with Annotated Narrationcited by 6
  30. 2024Debugging Defective Visualizations: Empirical Insights Informing a Human-AI Co-Debugging Systemcited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 24; 2026: 13.

37 papers.

Sort:

By area:

  1. 2025A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligencecited by 170
  2. 2025Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systemscited by 126
  3. 2025Atom of Thoughts for Markov LLM Test-Time Scalingcited by 54
  4. 2025MedicalAgentsBench for Complex Medical Reasoning: Comparing Internalized Reasoning Models versus Externalized Agent-based Frameworkscited by 40
  5. 2025AutoEnv: Automated Environments for Measuring Cross-Environment Agent Learningcited by 20
  6. 2025MEBench: Benchmarking Large Language Models for Cross-Document Multi-Entity Question Answeringcited by 16
  7. 2025Concise Reasoning, Big Gains: Pruning Long Reasoning Trace with Difficulty-Aware Promptingcited by 15
  8. 2025EllieSQL: Cost-Efficient Text-to-SQL with Complexity-Aware Routingcited by 13
  9. 2026Harnessing Agentic Evolutioncited by 11
  10. 2025LLM Prompt Duel Optimizer: Efficient Label-Free Prompt Optimizationcited by 8
  11. 2025SIPDO: Closed-Loop Prompt Optimization via Synthetic Data Feedbackcited by 7
  12. 2025No Loss, No Gain: Gated Refinement and Adaptive Compression for Prompt Optimizationcited by 7
25 more
  1. 2025Unlabeled Data Improves Fine-Grained Image Zero-shot Classification with Multimodal LLMscited by 6
  2. 2025Toward Automated Robustness Evaluation of Mathematical Reasoningcited by 6
  3. 2025MAPGD: Multi-Agent Prompt Gradient Descent for Collaborative Prompt Optimizationcited by 6
  4. 2025ZERA: Zero-init Instruction Evolving Refinement Agent -- From Zero Instructions to Structured Prompts via Principle-based Optimizationcited by 6
  5. 2026PathWise: Planning through World Model for Automated Heuristic Design via Self-Evolving LLMscited by 6
  6. 2025Explaining Large Language Models with gSMILEcited by 4
  7. 2025Code-enabled language models can outperform reasoning models on diverse taskscited by 4
  8. 2025PMPO: Probabilistic Metric Prompt Optimization for Small and Large Language Modelscited by 3
  9. 2025ORPP: Self-Optimizing Role-playing Prompts to Enhance Language Model Capabilitiescited by 3
  10. 2026JudgeFlow: Agentic Workflow Optimization via Block Judgecited by 3
  11. 2026Training LLM Agents for Spontaneous, Reward-Free Self-Evolution via World Knowledge Explorationcited by 3
  12. 2025Silence is Not Consensus: Disrupting Agreement Bias in Multi-Agent LLMs via Catfish Agent for Clinical Decision Makingcited by 2
  13. 2025Efficient Prompt Optimisation for Legal Text Classification with Proxy Prompt Evaluatorcited by 2
  14. 2026Scalable Environments Drive Generalizable Agentscited by 2
  15. 2025Rethinking Prompt Optimizers: From Prompt Merits to Optimizationcited by 1
  16. 2025MA-SAPO: Multi-Agent Reasoning for Score-Aware Prompt Optimizationcited by 1
  17. 2026UPA: Unsupervised Prompt Agent via Tree-Based Search and Selectioncited by 1
  18. 2026ArtifactLens: Hundreds of Labels Are Enough for Artifact Detection with VLMscited by 1
  19. 2025ELPO: Ensemble Learning Based Prompt Optimization for Large Language Modelscited by 0
  20. 2026C-MOP: Integrating Momentum and Boundary-Aware Clustering for Enhanced Prompt Evolutioncited by 0
  21. 2026RASPRef: Retrieval-Augmented Self-Supervised Prompt Refinement for Large Reasoning Modelscited by 0
  22. 2026A Systematic Approach for Large Language Models Debuggingcited by 0
  23. 2026Small Language Model Helps Resolve Semantic Ambiguity of LLM Promptcited by 0
  24. 2026Reinforced Collaboration in Multi-Agent Flow Networkscited by 0
  25. 2026Local Edits, Global Ripples: Replay-Informed Policy Adaptation for Workflow Synthesiscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024GReaTer: Gradients over Reasoning Makes Smaller Language Models Strong Prompt Optimizers
  2. 2025Local Prompt Optimization
  3. 2026Modular Prompt Optimization: Optimizing Structured Prompts with Section-Local Textual Gradients
  4. 2026Prompt Segmentation and Annotation Optimisation: Controlling LLM Behaviour via Optimised Segment-Level Annotations
  5. 2025Better by Comparison: Retrieval-Augmented Contrastive Reasoning for Automatic Prompt Optimization
  6. 2026Reflective Prompt Tuning through Language Model Function-Calling
  7. 2025PromptCoT 2.0: Scaling Prompt Synthesis for Large Language Model Reasoning
  8. 2026CAPO: Constraint-Aware Prompt Optimization for LLM Agents