{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:RGWEH4FGVULA7CDD7E433OUL65","short_pith_number":"pith:RGWEH4FG","canonical_record":{"source":{"id":"2508.07407","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-10T16:07:32Z","cross_cats_sorted":["cs.CL","cs.MA"],"title_canon_sha256":"cfd55accbaf5ef516bb178e65117a2080703b8c5ed092eedd40834f0e30ae4b2","abstract_canon_sha256":"55ad295007f5f557520b869117bfd72bd9bafa05ae801cf19492c71eedd4c90c"},"schema_version":"1.0"},"canonical_sha256":"89ac43f0a6ad160f8863f939bdba8bf77c45fcc3c05cde2e14e48be26f2ab773","source":{"kind":"arxiv","id":"2508.07407","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.07407","created_at":"2026-05-17T23:38:49Z"},{"alias_kind":"arxiv_version","alias_value":"2508.07407v2","created_at":"2026-05-17T23:38:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.07407","created_at":"2026-05-17T23:38:49Z"},{"alias_kind":"pith_short_12","alias_value":"RGWEH4FGVULA","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"RGWEH4FGVULA7CDD","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"RGWEH4FG","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:RGWEH4FGVULA7CDD7E433OUL65","target":"record","payload":{"canonical_record":{"source":{"id":"2508.07407","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-10T16:07:32Z","cross_cats_sorted":["cs.CL","cs.MA"],"title_canon_sha256":"cfd55accbaf5ef516bb178e65117a2080703b8c5ed092eedd40834f0e30ae4b2","abstract_canon_sha256":"55ad295007f5f557520b869117bfd72bd9bafa05ae801cf19492c71eedd4c90c"},"schema_version":"1.0"},"canonical_sha256":"89ac43f0a6ad160f8863f939bdba8bf77c45fcc3c05cde2e14e48be26f2ab773","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:49.802512Z","signature_b64":"UcNh8BTqJ4W5oC3QX7+OllMpVmDa8IZOXLUnfL5YRx78TUWXImbg6MXxxkN8fn4Ip3jEgO7iCAomIyUHEb83Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"89ac43f0a6ad160f8863f939bdba8bf77c45fcc3c05cde2e14e48be26f2ab773","last_reissued_at":"2026-05-17T23:38:49.801954Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:49.801954Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.07407","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:38:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rJHt8T592DHMEq2VnPmP13e9q6UMojC4tY3X/QE9KUkMeOfOJ+Ucz0VfuIcBfw5214Pe0KudBBMf4Frz2AJPBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-03T07:29:30.054821Z"},"content_sha256":"4dbb06bdb966025b05c6e0925fa42f548761fde7a97aed1720816fa1a4b8210c","schema_version":"1.0","event_id":"sha256:4dbb06bdb966025b05c6e0925fa42f548761fde7a97aed1720816fa1a4b8210c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:RGWEH4FGVULA7CDD7E433OUL65","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Self-evolving AI agents use interaction feedback to continuously improve beyond their initial foundation model capabilities.","cross_cats":["cs.CL","cs.MA"],"primary_cat":"cs.AI","authors_text":"Bin Wu, Guibin Zhang, Han Zhou, Jinyuan Fang, Nikos Aletras, Siwei Liu, Xinhao Yi, Xi Wang, Xi Zhang, Yanwen Peng, Yingxu Wang, Yi Xu, Zaiqiao Meng, Zhaochun Ren, Zihao Li","submitted_at":"2025-08-10T16:07:32Z","abstract_excerpt":"Recent advances in large language models have sparked growing interest in AI agents capable of solving complex, real-world tasks. However, most existing agent systems rely on manually crafted configurations that remain static after deployment, limiting their ability to adapt to dynamic and evolving environments. To this end, recent research has explored agent evolution techniques that aim to automatically enhance agent systems based on interaction data and environmental feedback. This emerging direction lays the foundation for self-evolving AI agents, which bridge the static capabilities of fo"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Self-evolving AI agents bridge the static capabilities of foundation models with the continuous adaptability required by lifelong agentic systems.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That existing agent evolution techniques can be meaningfully unified and compared under a single four-component framework of System Inputs, Agent System, Environment, and Optimisers.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A comprehensive review of self-evolving AI agents that improve themselves over time, organized via a framework of inputs, agent system, environment, and optimizers, with domain-specific and safety discussions.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Self-evolving AI agents use interaction feedback to continuously improve beyond their initial foundation model capabilities.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"9175afba36a5000dc4c69ece5fe86241d3d21dd94713adbaa65bad9050f422f9"},"source":{"id":"2508.07407","kind":"arxiv","version":2},"verdict":{"id":"70df4da8-fbae-4e08-a538-9db19d253ad1","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T23:17:51.188719Z","strongest_claim":"Self-evolving AI agents bridge the static capabilities of foundation models with the continuous adaptability required by lifelong agentic systems.","one_line_summary":"A comprehensive review of self-evolving AI agents that improve themselves over time, organized via a framework of inputs, agent system, environment, and optimizers, with domain-specific and safety discussions.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That existing agent evolution techniques can be meaningfully unified and compared under a single four-component framework of System Inputs, Agent System, Environment, and Optimisers.","pith_extraction_headline":"Self-evolving AI agents use interaction feedback to continuously improve beyond their initial foundation model capabilities."},"references":{"count":126,"sample":[{"doi":"","year":null,"title":"arXiv preprint arXiv:2502.02928 , year=","work_id":"418dca89-6599-471b-85c5-d854ea2413d2","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"PromptWizard: Task-aware prompt optimization framework","work_id":"62e6ddc6-6ade-445a-877d-f5717ca72c59","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"arXiv preprint arXiv:2503.22678 , year=","work_id":"4f8052bb-827e-4e6d-9cbd-6ea7f9de635e","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"arXiv preprint arXiv:2402.09015 , year=","work_id":"00aca99a-c7d8-4c3c-9108-d9ec0bfc153b","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"34 Derek Austin and Elliott Chartock.GRAD-SUM: Leveraging gradient summarization for optimal prompt engineering. arXiv preprint arXiv:2407.12865,","work_id":"c6c9c1a9-b58d-4836-bfc4-40cf7cce67a3","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":126,"snapshot_sha256":"902a489073d04f4bf489b12623971ad7d27e47d76cf8d950aaa9d46f8283dea4","internal_anchors":25},"formal_canon":{"evidence_count":2,"snapshot_sha256":"6280d44f686ecddeffc840fcf25d889890e4bffe5b9b66fcebc978b54f166379"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"70df4da8-fbae-4e08-a538-9db19d253ad1"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:38:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"EX7Cp6brLkErT5D8fwTsxXWVMp7E08dbvHBS0cDKBPmXMUzI1jU/Nm7jLK5ZzG+QShkHLK0QeuaazWR01dCjAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-03T07:29:30.055357Z"},"content_sha256":"ffd4d564a519cdb8b83e1fbb0351d1b2d85e3ad67a8640a208c86adfa9b486d8","schema_version":"1.0","event_id":"sha256:ffd4d564a519cdb8b83e1fbb0351d1b2d85e3ad67a8640a208c86adfa9b486d8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RGWEH4FGVULA7CDD7E433OUL65/bundle.json","state_url":"https://pith.science/pith/RGWEH4FGVULA7CDD7E433OUL65/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RGWEH4FGVULA7CDD7E433OUL65/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-03T07:29:30Z","links":{"resolver":"https://pith.science/pith/RGWEH4FGVULA7CDD7E433OUL65","bundle":"https://pith.science/pith/RGWEH4FGVULA7CDD7E433OUL65/bundle.json","state":"https://pith.science/pith/RGWEH4FGVULA7CDD7E433OUL65/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RGWEH4FGVULA7CDD7E433OUL65/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:RGWEH4FGVULA7CDD7E433OUL65","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"55ad295007f5f557520b869117bfd72bd9bafa05ae801cf19492c71eedd4c90c","cross_cats_sorted":["cs.CL","cs.MA"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-10T16:07:32Z","title_canon_sha256":"cfd55accbaf5ef516bb178e65117a2080703b8c5ed092eedd40834f0e30ae4b2"},"schema_version":"1.0","source":{"id":"2508.07407","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.07407","created_at":"2026-05-17T23:38:49Z"},{"alias_kind":"arxiv_version","alias_value":"2508.07407v2","created_at":"2026-05-17T23:38:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.07407","created_at":"2026-05-17T23:38:49Z"},{"alias_kind":"pith_short_12","alias_value":"RGWEH4FGVULA","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"RGWEH4FGVULA7CDD","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"RGWEH4FG","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:ffd4d564a519cdb8b83e1fbb0351d1b2d85e3ad67a8640a208c86adfa9b486d8","target":"graph","created_at":"2026-05-17T23:38:49Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Self-evolving AI agents bridge the static capabilities of foundation models with the continuous adaptability required by lifelong agentic systems."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That existing agent evolution techniques can be meaningfully unified and compared under a single four-component framework of System Inputs, Agent System, Environment, and Optimisers."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"A comprehensive review of self-evolving AI agents that improve themselves over time, organized via a framework of inputs, agent system, environment, and optimizers, with domain-specific and safety discussions."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Self-evolving AI agents use interaction feedback to continuously improve beyond their initial foundation model capabilities."}],"snapshot_sha256":"9175afba36a5000dc4c69ece5fe86241d3d21dd94713adbaa65bad9050f422f9"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"6280d44f686ecddeffc840fcf25d889890e4bffe5b9b66fcebc978b54f166379"},"paper":{"abstract_excerpt":"Recent advances in large language models have sparked growing interest in AI agents capable of solving complex, real-world tasks. However, most existing agent systems rely on manually crafted configurations that remain static after deployment, limiting their ability to adapt to dynamic and evolving environments. To this end, recent research has explored agent evolution techniques that aim to automatically enhance agent systems based on interaction data and environmental feedback. This emerging direction lays the foundation for self-evolving AI agents, which bridge the static capabilities of fo","authors_text":"Bin Wu, Guibin Zhang, Han Zhou, Jinyuan Fang, Nikos Aletras, Siwei Liu, Xinhao Yi, Xi Wang, Xi Zhang, Yanwen Peng, Yingxu Wang, Yi Xu, Zaiqiao Meng, Zhaochun Ren, Zihao Li","cross_cats":["cs.CL","cs.MA"],"headline":"Self-evolving AI agents use interaction feedback to continuously improve beyond their initial foundation model capabilities.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-10T16:07:32Z","title":"A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems"},"references":{"count":126,"internal_anchors":25,"resolved_work":126,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"arXiv preprint arXiv:2502.02928 , year=","work_id":"418dca89-6599-471b-85c5-d854ea2413d2","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"PromptWizard: Task-aware prompt optimization framework","work_id":"62e6ddc6-6ade-445a-877d-f5717ca72c59","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"arXiv preprint arXiv:2503.22678 , year=","work_id":"4f8052bb-827e-4e6d-9cbd-6ea7f9de635e","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"arXiv preprint arXiv:2402.09015 , year=","work_id":"00aca99a-c7d8-4c3c-9108-d9ec0bfc153b","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"34 Derek Austin and Elliott Chartock.GRAD-SUM: Leveraging gradient summarization for optimal prompt engineering. arXiv preprint arXiv:2407.12865,","work_id":"c6c9c1a9-b58d-4836-bfc4-40cf7cce67a3","year":null}],"snapshot_sha256":"902a489073d04f4bf489b12623971ad7d27e47d76cf8d950aaa9d46f8283dea4"},"source":{"id":"2508.07407","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-15T23:17:51.188719Z","id":"70df4da8-fbae-4e08-a538-9db19d253ad1","model_set":{"reader":"grok-4.3"},"one_line_summary":"A comprehensive review of self-evolving AI agents that improve themselves over time, organized via a framework of inputs, agent system, environment, and optimizers, with domain-specific and safety discussions.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Self-evolving AI agents use interaction feedback to continuously improve beyond their initial foundation model capabilities.","strongest_claim":"Self-evolving AI agents bridge the static capabilities of foundation models with the continuous adaptability required by lifelong agentic systems.","weakest_assumption":"That existing agent evolution techniques can be meaningfully unified and compared under a single four-component framework of System Inputs, Agent System, Environment, and Optimisers."}},"verdict_id":"70df4da8-fbae-4e08-a538-9db19d253ad1"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:4dbb06bdb966025b05c6e0925fa42f548761fde7a97aed1720816fa1a4b8210c","target":"record","created_at":"2026-05-17T23:38:49Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"55ad295007f5f557520b869117bfd72bd9bafa05ae801cf19492c71eedd4c90c","cross_cats_sorted":["cs.CL","cs.MA"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-10T16:07:32Z","title_canon_sha256":"cfd55accbaf5ef516bb178e65117a2080703b8c5ed092eedd40834f0e30ae4b2"},"schema_version":"1.0","source":{"id":"2508.07407","kind":"arxiv","version":2}},"canonical_sha256":"89ac43f0a6ad160f8863f939bdba8bf77c45fcc3c05cde2e14e48be26f2ab773","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"89ac43f0a6ad160f8863f939bdba8bf77c45fcc3c05cde2e14e48be26f2ab773","first_computed_at":"2026-05-17T23:38:49.801954Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:38:49.801954Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"UcNh8BTqJ4W5oC3QX7+OllMpVmDa8IZOXLUnfL5YRx78TUWXImbg6MXxxkN8fn4Ip3jEgO7iCAomIyUHEb83Dg==","signature_status":"signed_v1","signed_at":"2026-05-17T23:38:49.802512Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.07407","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4dbb06bdb966025b05c6e0925fa42f548761fde7a97aed1720816fa1a4b8210c","sha256:ffd4d564a519cdb8b83e1fbb0351d1b2d85e3ad67a8640a208c86adfa9b486d8"],"state_sha256":"91f7119222aea4c0d5c86116607c3ab0b3d377271447e8852531fb11bcef2965"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZOJz54YRIDnyqAzOVEPD09YU5cnBfhIyYATiiBPxTU6jryi1qL+EcAB+OJH1epMqZ0/PGBitqqEjdg4p2dy4CA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-03T07:29:30.057803Z","bundle_sha256":"9ce4d24c1f3b16e8e0f7ae7ac6551a5cf7905f7f39c34abd01f71bf7a9bf9483"}}