{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:UQOLUGOKYGIVAULY22FW7O2EDV","short_pith_number":"pith:UQOLUGOK","schema_version":"1.0","canonical_sha256":"a41cba19cac191505178d68b6fbb441d69e044317dd655aadab9caa126e745e9","source":{"kind":"arxiv","id":"2507.21046","version":4},"attestation_state":"computed","paper":{"title":"A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Self-evolving agents adapt their internal components through ongoing interactions to move beyond static large language models toward artificial super intelligence.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Cheng Qian, Dongrui Liu, Han Xiao, Heng Ji, Hongru Wang, Hongzhang Liu, Huan-ang Gao, Huazheng Wang, Jiahao Qiu, Jiayi Geng, Jiayi Zhang, Jinyu Xiang, Mengdi Wang, Mengkang Hu, Minda Hu, Qihan Ren, Qingyun Wu, Qiwen Zhao, Shaokun Zhang, Shilong Liu, Wenyue Hua, Xinzhe Juan, Xuan Qi, Yiran Wu, Yixiong Fang, Yuhang Zhou, Zhenhailong Wang","submitted_at":"2025-07-28T17:59:05Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated remarkable capabilities across diverse tasks but remain fundamentally static, unable to adapt their internal parameters to novel tasks, evolving knowledge domains, or dynamic interaction contexts. As LLMs are increasingly deployed in open-ended, interactive environments, this static nature has become a critical bottleneck, necessitating agents that can adaptively reason, act, and evolve in real time. This paradigm shift -- from scaling static models to developing self-evolving agents -- has sparked growing interest in architectures and methods ena"},"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":"2507.21046","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-07-28T17:59:05Z","cross_cats_sorted":[],"title_canon_sha256":"5fd67d2ad00a3537bae407d414a8c2d5d4da0b3fb646347d6e313a0d26a0014f","abstract_canon_sha256":"5e095e3391867160653090fa894da636081a6f3c49d79f60df93efe581ee041d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:21.476475Z","signature_b64":"ZymMJefzpA6aNV6qfbx0pEPUAphTnpyPe4g5z6T9krS639N6iTC1paIT4gEMyiRb6r6CoF/8Whp+HnJiNLedCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a41cba19cac191505178d68b6fbb441d69e044317dd655aadab9caa126e745e9","last_reissued_at":"2026-05-17T23:39:21.475689Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:21.475689Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Self-evolving agents adapt their internal components through ongoing interactions to move beyond static large language models toward artificial super intelligence.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Cheng Qian, Dongrui Liu, Han Xiao, Heng Ji, Hongru Wang, Hongzhang Liu, Huan-ang Gao, Huazheng Wang, Jiahao Qiu, Jiayi Geng, Jiayi Zhang, Jinyu Xiang, Mengdi Wang, Mengkang Hu, Minda Hu, Qihan Ren, Qingyun Wu, Qiwen Zhao, Shaokun Zhang, Shilong Liu, Wenyue Hua, Xinzhe Juan, Xuan Qi, Yiran Wu, Yixiong Fang, Yuhang Zhou, Zhenhailong Wang","submitted_at":"2025-07-28T17:59:05Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated remarkable capabilities across diverse tasks but remain fundamentally static, unable to adapt their internal parameters to novel tasks, evolving knowledge domains, or dynamic interaction contexts. As LLMs are increasingly deployed in open-ended, interactive environments, this static nature has become a critical bottleneck, necessitating agents that can adaptively reason, act, and evolve in real time. This paradigm shift -- from scaling static models to developing self-evolving agents -- has sparked growing interest in architectures and methods ena"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"This survey provides the first systematic and comprehensive review of self-evolving agents, organizing the field around three foundational dimensions: what, when, and how to evolve.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The forward-looking premise that self-evolving agents constitute the primary route to Artificial Super Intelligence, which rests on an unproven long-term vision rather than demonstrated evidence.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"The paper delivers the first systematic review of self-evolving agents, structured around what components evolve, when adaptation occurs, and how it is implemented.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Self-evolving agents adapt their internal components through ongoing interactions to move beyond static large language models toward artificial super intelligence.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"9396312e5000b5677efe5bef695002ad69292cb811f67170f070cc3d83cdb3f3"},"source":{"id":"2507.21046","kind":"arxiv","version":4},"verdict":{"id":"99375796-78d2-4497-b9a3-1f30ebd8f417","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T22:18:23.448291Z","strongest_claim":"This survey provides the first systematic and comprehensive review of self-evolving agents, organizing the field around three foundational dimensions: what, when, and how to evolve.","one_line_summary":"The paper delivers the first systematic review of self-evolving agents, structured around what components evolve, when adaptation occurs, and how it is implemented.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The forward-looking premise that self-evolving agents constitute the primary route to Artificial Super Intelligence, which rests on an unproven long-term vision rather than demonstrated evidence.","pith_extraction_headline":"Self-evolving agents adapt their internal components through ongoing interactions to move beyond static large language models toward artificial super intelligence."},"references":{"count":297,"sample":[{"doi":"","year":2025,"title":"Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems , author=. 2025 , eprint=","work_id":"cd53ae35-6014-474a-87ab-4121642fa3d7","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1145/3626772.3661381","year":2024,"title":"ISBN 9798400704314","work_id":"f7cdc348-9f85-4c96-817c-918f5b50df36","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Toward a Theory of Agents as Tool-Use Decision-Makers , author=. 2025 , eprint=","work_id":"18fe3124-9c04-4c1a-8c4a-fade4bb0cd5f","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1007/s11704-024-40231-1","year":null,"title":"A survey on large language model based autonomous agents","work_id":"c00808e5-64e6-4bbb-9a05-230c05403bff","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Large Language Model Agent: A Survey on Methodology, Applications and Challenges , author=. 2025 , eprint=","work_id":"4cb9fbbc-6731-4272-8bfa-cdb9bdf15c9d","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":297,"snapshot_sha256":"9398d237834539b76e9db757193a1d1b4b259ce4b2e9f7474fe08cae9788ed0d","internal_anchors":47},"formal_canon":{"evidence_count":3,"snapshot_sha256":"4b7f78df14b56f2fba9b11f18d0694bb86792900b887ed9a026f0e63c81afbb6"},"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":"2507.21046","created_at":"2026-05-17T23:39:21.475837+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.21046v4","created_at":"2026-05-17T23:39:21.475837+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.21046","created_at":"2026-05-17T23:39:21.475837+00:00"},{"alias_kind":"pith_short_12","alias_value":"UQOLUGOKYGIV","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"UQOLUGOKYGIVAULY","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"UQOLUGOK","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":48,"internal_anchor_count":48,"sample":[{"citing_arxiv_id":"2605.21951","citing_title":"Dynamic Mixture of Latent Memories for Self-Evolving Agents","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"2510.16079","citing_title":"EvolveR: Self-Evolving LLM Agents through an Experience-Driven Lifecycle","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2602.06470","citing_title":"Improve Large Language Model Systems with User Logs","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2605.20189","citing_title":"SOLAR: A Self-Optimizing Open-Ended Autonomous Agent for Lifelong Learning and Continual Adaptation","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2604.15034","citing_title":"Autogenesis: A Self-Evolving Agent Protocol","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14133","citing_title":"ClawForge: Generating Executable Interactive Benchmarks for Command-Line Agents","ref_index":47,"is_internal_anchor":true},{"citing_arxiv_id":"2605.16508","citing_title":"The Scaling Laws of Skills in LLM Agent Systems","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18181","citing_title":"Scalable Environments Drive Generalizable Agents","ref_index":10,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18930","citing_title":"OEP: Poisoning Self-Evolving LLM Agents via Locally Correct but Non-Transferable Experiences","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2604.27859","citing_title":"Rethinking Agentic Reinforcement Learning In Large Language Models","ref_index":22,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14892","citing_title":"Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systems","ref_index":31,"is_internal_anchor":true},{"citing_arxiv_id":"2605.15384","citing_title":"Is One Score Enough? Rethinking the Evaluation of Sequentially Evolving LLM Memory","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2509.02547","citing_title":"The Landscape of Agentic Reinforcement Learning for LLMs: A Survey","ref_index":151,"is_internal_anchor":true},{"citing_arxiv_id":"2510.16079","citing_title":"EvolveR: Self-Evolving LLM Agents through an Experience-Driven Lifecycle","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2510.23883","citing_title":"Agentic AI Security: Threats, Defenses, Evaluation, and Open Challenges","ref_index":52,"is_internal_anchor":true},{"citing_arxiv_id":"2601.12538","citing_title":"Agentic Reasoning for Large Language Models","ref_index":59,"is_internal_anchor":true},{"citing_arxiv_id":"2511.21678","citing_title":"Agentic Learner with Grow-and-Refine Multimodal Semantic Memory","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2512.13399","citing_title":"Differentiable Evolutionary Reinforcement Learning","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2512.15567","citing_title":"Evaluating Large Language Models in Scientific Discovery","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2602.07900","citing_title":"Rethinking the Value of Agent-Generated Tests for LLM-Based Software Engineering Agents","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2604.09579","citing_title":"Help Without Being Asked: A Deployed Proactive Agent System for On-Call Support with Continuous Self-Improvement","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2604.16322","citing_title":"Steerable Instruction Following Coding Data Synthesis with Actor-Parametric Schema Co-Evolution","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14133","citing_title":"ClawForge: Generating Executable Interactive Benchmarks for Command-Line Agents","ref_index":47,"is_internal_anchor":true},{"citing_arxiv_id":"2605.13941","citing_title":"EvolveMem:Self-Evolving Memory Architecture via AutoResearch for LLM Agents","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14892","citing_title":"Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systems","ref_index":30,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":3,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UQOLUGOKYGIVAULY22FW7O2EDV","json":"https://pith.science/pith/UQOLUGOKYGIVAULY22FW7O2EDV.json","graph_json":"https://pith.science/api/pith-number/UQOLUGOKYGIVAULY22FW7O2EDV/graph.json","events_json":"https://pith.science/api/pith-number/UQOLUGOKYGIVAULY22FW7O2EDV/events.json","paper":"https://pith.science/paper/UQOLUGOK"},"agent_actions":{"view_html":"https://pith.science/pith/UQOLUGOKYGIVAULY22FW7O2EDV","download_json":"https://pith.science/pith/UQOLUGOKYGIVAULY22FW7O2EDV.json","view_paper":"https://pith.science/paper/UQOLUGOK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.21046&json=true","fetch_graph":"https://pith.science/api/pith-number/UQOLUGOKYGIVAULY22FW7O2EDV/graph.json","fetch_events":"https://pith.science/api/pith-number/UQOLUGOKYGIVAULY22FW7O2EDV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UQOLUGOKYGIVAULY22FW7O2EDV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UQOLUGOKYGIVAULY22FW7O2EDV/action/storage_attestation","attest_author":"https://pith.science/pith/UQOLUGOKYGIVAULY22FW7O2EDV/action/author_attestation","sign_citation":"https://pith.science/pith/UQOLUGOKYGIVAULY22FW7O2EDV/action/citation_signature","submit_replication":"https://pith.science/pith/UQOLUGOKYGIVAULY22FW7O2EDV/action/replication_record"}},"created_at":"2026-05-17T23:39:21.475837+00:00","updated_at":"2026-05-17T23:39:21.475837+00:00"}