{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:QOBYKAH63XST3RJGJYHFW4EZVN","short_pith_number":"pith:QOBYKAH6","schema_version":"1.0","canonical_sha256":"83838500fedde53dc5264e0e5b7099ab7949e247912d715b5417e9465f33d95a","source":{"kind":"arxiv","id":"2607.05202","version":1},"attestation_state":"computed","paper":{"title":"EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Chuanrui Hu, Hongda Chen, Jie Gui, Pengfei Yao, Teng Li, Xiaofeng Cong, Xingze Gao, Yafeng Deng, Yi Bai, Yunyun Han, Zhao Wang, Zhengwei Wu","submitted_at":"2026-07-06T15:17:09Z","abstract_excerpt":"Agent self-evolution in long-horizon LLM systems is largely procedural: useful experience is not merely stored information, but reusable procedures for searching, debugging, and verification. Yet current evaluations do not isolate this form of transfer. Agent benchmarks test single-episode task solving; memory benchmarks target information retention rather than procedural reuse. We introduce EvoAgentBench, a benchmark for agent self-evolution via Ability-guided transfer across four agentic domains: web research, algorithmic reasoning, software engineering, and knowledge work. EvoAgentBench ext"},"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":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2607.05202","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-06T15:17:09Z","cross_cats_sorted":[],"title_canon_sha256":"b74d232ff18eb60315dcf049ce73ab6a60c7f7a89370685ed70475ffdccf2370","abstract_canon_sha256":"2fabe4b9e198373ba5c5e93e2e2d55e167e476e889061c09b3fce24dc592ceca"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T03:19:19.627822Z","signature_b64":"Y58SUjADPpBUQ08slf30+6/uDH5c05qkIx+dQlnLJ8IL0F3mrh0D3RXLjdJnV4NR3SQm6oR7Gr+MWjv6tDRJDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"83838500fedde53dc5264e0e5b7099ab7949e247912d715b5417e9465f33d95a","last_reissued_at":"2026-07-07T03:19:19.627353Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T03:19:19.627353Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Chuanrui Hu, Hongda Chen, Jie Gui, Pengfei Yao, Teng Li, Xiaofeng Cong, Xingze Gao, Yafeng Deng, Yi Bai, Yunyun Han, Zhao Wang, Zhengwei Wu","submitted_at":"2026-07-06T15:17:09Z","abstract_excerpt":"Agent self-evolution in long-horizon LLM systems is largely procedural: useful experience is not merely stored information, but reusable procedures for searching, debugging, and verification. Yet current evaluations do not isolate this form of transfer. Agent benchmarks test single-episode task solving; memory benchmarks target information retention rather than procedural reuse. We introduce EvoAgentBench, a benchmark for agent self-evolution via Ability-guided transfer across four agentic domains: web research, algorithmic reasoning, software engineering, and knowledge work. EvoAgentBench ext"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.05202","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2607.05202/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"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":"2607.05202","created_at":"2026-07-07T03:19:19.627434+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.05202v1","created_at":"2026-07-07T03:19:19.627434+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.05202","created_at":"2026-07-07T03:19:19.627434+00:00"},{"alias_kind":"pith_short_12","alias_value":"QOBYKAH63XST","created_at":"2026-07-07T03:19:19.627434+00:00"},{"alias_kind":"pith_short_16","alias_value":"QOBYKAH63XST3RJG","created_at":"2026-07-07T03:19:19.627434+00:00"},{"alias_kind":"pith_short_8","alias_value":"QOBYKAH6","created_at":"2026-07-07T03:19:19.627434+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2608.03764","citing_title":"GDPevo: Evaluating Agent Self-Evolution on Real Business Tasks","ref_index":10,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QOBYKAH63XST3RJGJYHFW4EZVN","json":"https://pith.science/pith/QOBYKAH63XST3RJGJYHFW4EZVN.json","graph_json":"https://pith.science/api/pith-number/QOBYKAH63XST3RJGJYHFW4EZVN/graph.json","events_json":"https://pith.science/api/pith-number/QOBYKAH63XST3RJGJYHFW4EZVN/events.json","paper":"https://pith.science/paper/QOBYKAH6"},"agent_actions":{"view_html":"https://pith.science/pith/QOBYKAH63XST3RJGJYHFW4EZVN","download_json":"https://pith.science/pith/QOBYKAH63XST3RJGJYHFW4EZVN.json","view_paper":"https://pith.science/paper/QOBYKAH6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.05202&json=true","fetch_graph":"https://pith.science/api/pith-number/QOBYKAH63XST3RJGJYHFW4EZVN/graph.json","fetch_events":"https://pith.science/api/pith-number/QOBYKAH63XST3RJGJYHFW4EZVN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QOBYKAH63XST3RJGJYHFW4EZVN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QOBYKAH63XST3RJGJYHFW4EZVN/action/storage_attestation","attest_author":"https://pith.science/pith/QOBYKAH63XST3RJGJYHFW4EZVN/action/author_attestation","sign_citation":"https://pith.science/pith/QOBYKAH63XST3RJGJYHFW4EZVN/action/citation_signature","submit_replication":"https://pith.science/pith/QOBYKAH63XST3RJGJYHFW4EZVN/action/replication_record"}},"created_at":"2026-07-07T03:19:19.627434+00:00","updated_at":"2026-07-07T03:19:19.627434+00:00"}