{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:JMJJNPZBO6PRVNC2QRNEBSWN33","short_pith_number":"pith:JMJJNPZB","schema_version":"1.0","canonical_sha256":"4b1296bf21779f1ab45a845a40cacddeff4d6a06eb4c42057b7fc3ddd4e5c667","source":{"kind":"arxiv","id":"2309.16797","version":1},"attestation_state":"computed","paper":{"title":"Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution","license":"http://creativecommons.org/licenses/by/4.0/","headline":"An LLM can improve prompting by evolving both the task prompts and the mutation rules that generate them.","cross_cats":["cs.AI","cs.LG","cs.NE"],"primary_cat":"cs.CL","authors_text":"Chrisantha Fernando, Dylan Banarse, Henryk Michalewski, Simon Osindero, Tim Rockt\\\"aschel","submitted_at":"2023-09-28T19:01:07Z","abstract_excerpt":"Popular prompt strategies like Chain-of-Thought Prompting can dramatically improve the reasoning abilities of Large Language Models (LLMs) in various domains. However, such hand-crafted prompt-strategies are often sub-optimal. In this paper, we present Promptbreeder, a general-purpose self-referential self-improvement mechanism that evolves and adapts prompts for a given domain. Driven by an LLM, Promptbreeder mutates a population of task-prompts, and subsequently evaluates them for fitness on a training set. Crucially, the mutation of these task-prompts is governed by mutation-prompts that th"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2309.16797","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-09-28T19:01:07Z","cross_cats_sorted":["cs.AI","cs.LG","cs.NE"],"title_canon_sha256":"4244f6eb42bf5cfaa0a8ca2bc7803a3815d151f208b15816561832a5396f68b0","abstract_canon_sha256":"72442a4596cb224cd8ca06a06d4d596f0073c6c055ab633d35e6e856a0f40a39"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:48.546255Z","signature_b64":"8T7YanHoer+tfURCPgaHhzLNVbqg7q4VHvsz9dGqwnv6l3OCe537XBVMQZ9CZvhBEt2ILmrZThH0i5nm54yLAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4b1296bf21779f1ab45a845a40cacddeff4d6a06eb4c42057b7fc3ddd4e5c667","last_reissued_at":"2026-05-17T23:38:48.545777Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:48.545777Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution","license":"http://creativecommons.org/licenses/by/4.0/","headline":"An LLM can improve prompting by evolving both the task prompts and the mutation rules that generate them.","cross_cats":["cs.AI","cs.LG","cs.NE"],"primary_cat":"cs.CL","authors_text":"Chrisantha Fernando, Dylan Banarse, Henryk Michalewski, Simon Osindero, Tim Rockt\\\"aschel","submitted_at":"2023-09-28T19:01:07Z","abstract_excerpt":"Popular prompt strategies like Chain-of-Thought Prompting can dramatically improve the reasoning abilities of Large Language Models (LLMs) in various domains. However, such hand-crafted prompt-strategies are often sub-optimal. In this paper, we present Promptbreeder, a general-purpose self-referential self-improvement mechanism that evolves and adapts prompts for a given domain. Driven by an LLM, Promptbreeder mutates a population of task-prompts, and subsequently evaluates them for fitness on a training set. Crucially, the mutation of these task-prompts is governed by mutation-prompts that th"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Promptbreeder outperforms state-of-the-art prompt strategies such as Chain-of-Thought and Plan-and-Solve Prompting on commonly used arithmetic and commonsense reasoning benchmarks. Furthermore, Promptbreeder is able to evolve intricate task-prompts for the challenging problem of hate speech classification.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the LLM can generate useful mutations and provide reliable fitness evaluations on a training set without systematic biases or errors that would derail the evolutionary process.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Promptbreeder evolves both task prompts and the mutation prompts that improve them using LLMs, outperforming Chain-of-Thought and Plan-and-Solve on arithmetic and commonsense reasoning benchmarks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"An LLM can improve prompting by evolving both the task prompts and the mutation rules that generate them.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"af8a4608b93492c9a639316a2a8363210f11448aa1c8d3fcc9d2985291fd12a3"},"source":{"id":"2309.16797","kind":"arxiv","version":1},"verdict":{"id":"7356eb50-875e-4a35-873b-bf88b81cd52d","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T08:08:25.069041Z","strongest_claim":"Promptbreeder outperforms state-of-the-art prompt strategies such as Chain-of-Thought and Plan-and-Solve Prompting on commonly used arithmetic and commonsense reasoning benchmarks. Furthermore, Promptbreeder is able to evolve intricate task-prompts for the challenging problem of hate speech classification.","one_line_summary":"Promptbreeder evolves both task prompts and the mutation prompts that improve them using LLMs, outperforming Chain-of-Thought and Plan-and-Solve on arithmetic and commonsense reasoning benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the LLM can generate useful mutations and provide reliable fitness evaluations on a training set without systematic biases or errors that would derail the evolutionary process.","pith_extraction_headline":"An LLM can improve prompting by evolving both the task prompts and the mutation rules that generate them."},"references":{"count":296,"sample":[{"doi":"","year":2021,"title":"Show Your Work: Scratchpads for Intermediate Computation with Language Models","work_id":"a05b1e60-8e76-4f26-9bea-28927a5f8620","ref_index":1,"cited_arxiv_id":"2112.00114","is_internal_anchor":true},{"doi":"","year":1995,"title":"The Hitchhiker's Guide to the Galaxy , author=. 1995 , publisher=","work_id":"07683f0c-cb34-47d6-83ae-f5d0726ac43a","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"NeurIPS , year =","work_id":"387c2ec4-3205-43fa-9107-bd3febe774bc","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"The Eleventh International Conference on Learning Representations,","work_id":"399e38b9-d994-4207-a188-550020e608cf","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"gradient descent","work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","ref_index":6,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":296,"snapshot_sha256":"db2d6045761ca65dfd1b0fc1282cac50449235add2e3e1a7df7c663909c2df89","internal_anchors":67},"formal_canon":{"evidence_count":2,"snapshot_sha256":"34973332bdf97d1893a8162d3a15016cc6de881333fbca73ea85f05de7f36b4e"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2309.16797","created_at":"2026-05-17T23:38:48.545857+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.16797v1","created_at":"2026-05-17T23:38:48.545857+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.16797","created_at":"2026-05-17T23:38:48.545857+00:00"},{"alias_kind":"pith_short_12","alias_value":"JMJJNPZBO6PR","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"JMJJNPZBO6PRVNC2","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"JMJJNPZB","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":32,"internal_anchor_count":32,"sample":[{"citing_arxiv_id":"2605.21792","citing_title":"Residual Skill Optimization for Text-to-SQL Ensembles","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2605.22166","citing_title":"Adapting the Interface, Not the Model: Runtime Harness Adaptation for Deterministic LLM Agents","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2605.21516","citing_title":"Harnesses for Inference-Time Alignment over Execution Trajectories","ref_index":31,"is_internal_anchor":true},{"citing_arxiv_id":"2605.21318","citing_title":"TextReg: Mitigating Prompt Distributional Overfitting via Regularized Text-Space Optimization","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2605.15665","citing_title":"PRISM: Prompt Reliability via Iterative Simulation and Monitoring for Enterprise Conversational AI","ref_index":10,"is_internal_anchor":true},{"citing_arxiv_id":"2605.15721","citing_title":"Contexting as Recommendation: Evolutionary Collaborative Filtering for Context Engineering","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2605.16233","citing_title":"FORGE: Self-Evolving Agent Memory With No Weight Updates via Population Broadcast","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2605.19093","citing_title":"Embedding by Elicitation: Dynamic Representations for Bayesian Optimization of System Prompts","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2605.19102","citing_title":"Prompt Optimization for LLM Code Generation via Reinforcement Learning","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2605.19633","citing_title":"optimize_anything: A Universal API for Optimizing any Text Parameter","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2605.17510","citing_title":"Scale-Dependent Collective Adaptation in Self-Amending LLM Societies: A Cross-Family Study of Emergent Governance","ref_index":36,"is_internal_anchor":true},{"citing_arxiv_id":"2512.08984","citing_title":"RAG-HAR: Retrieval Augmented Generation-based Human Activity Recognition","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2601.20981","citing_title":"Diversifying Toxicity Search in Large Language Models Through Speciation","ref_index":10,"is_internal_anchor":true},{"citing_arxiv_id":"2501.09686","citing_title":"Towards Large Reasoning Models: A Survey of Reinforced Reasoning with Large Language Models","ref_index":38,"is_internal_anchor":true},{"citing_arxiv_id":"2603.12510","citing_title":"Red-Teaming Vision-Language-Action Models via Quality Diversity Prompt Generation for Robust Robot Policies","ref_index":39,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12484","citing_title":"Learning, Fast and Slow: Towards LLMs That Adapt Continually","ref_index":15,"is_internal_anchor":true},{"citing_arxiv_id":"2309.03409","citing_title":"Large Language Models as Optimizers","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2507.21046","citing_title":"A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence","ref_index":41,"is_internal_anchor":true},{"citing_arxiv_id":"2605.13384","citing_title":"Teaching and Learning under Deductive Errors","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12484","citing_title":"Learning, Fast and Slow: Towards LLMs That Adapt Continually","ref_index":15,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08769","citing_title":"EvoMAS: Learning Execution-Time Workflows for Multi-Agent Systems","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10366","citing_title":"EGL-SCA: Structural Credit Assignment for Co-Evolving Instructions and Tools in Graph Reasoning Agents","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2604.23640","citing_title":"Prompt-Unknown Promotion Attacks against LLM-based Sequential Recommender Systems","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2605.04107","citing_title":"TSCG: Deterministic Tool-Schema Compilation for Agentic LLM Deployments","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2604.12601","citing_title":"LLM-Guided Prompt Evolution for Password Guessing","ref_index":5,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JMJJNPZBO6PRVNC2QRNEBSWN33","json":"https://pith.science/pith/JMJJNPZBO6PRVNC2QRNEBSWN33.json","graph_json":"https://pith.science/api/pith-number/JMJJNPZBO6PRVNC2QRNEBSWN33/graph.json","events_json":"https://pith.science/api/pith-number/JMJJNPZBO6PRVNC2QRNEBSWN33/events.json","paper":"https://pith.science/paper/JMJJNPZB"},"agent_actions":{"view_html":"https://pith.science/pith/JMJJNPZBO6PRVNC2QRNEBSWN33","download_json":"https://pith.science/pith/JMJJNPZBO6PRVNC2QRNEBSWN33.json","view_paper":"https://pith.science/paper/JMJJNPZB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.16797&json=true","fetch_graph":"https://pith.science/api/pith-number/JMJJNPZBO6PRVNC2QRNEBSWN33/graph.json","fetch_events":"https://pith.science/api/pith-number/JMJJNPZBO6PRVNC2QRNEBSWN33/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JMJJNPZBO6PRVNC2QRNEBSWN33/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JMJJNPZBO6PRVNC2QRNEBSWN33/action/storage_attestation","attest_author":"https://pith.science/pith/JMJJNPZBO6PRVNC2QRNEBSWN33/action/author_attestation","sign_citation":"https://pith.science/pith/JMJJNPZBO6PRVNC2QRNEBSWN33/action/citation_signature","submit_replication":"https://pith.science/pith/JMJJNPZBO6PRVNC2QRNEBSWN33/action/replication_record"}},"created_at":"2026-05-17T23:38:48.545857+00:00","updated_at":"2026-05-17T23:38:48.545857+00:00"}