{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:F4EGOHLGEIRXKH7D3BSNE2FFSG","short_pith_number":"pith:F4EGOHLG","schema_version":"1.0","canonical_sha256":"2f08671d662223751fe3d864d268a59197451741bb0d370e94a5a925f2451812","source":{"kind":"arxiv","id":"2605.29225","version":1},"attestation_state":"computed","paper":{"title":"BenchTrace: A Benchmark for Testing Reflection Ability and Controlled Evolution in LLM Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Akiko Aizawa, Fei Cheng, Jiahao Huang, Junfeng Jiang, Zefan Yu","submitted_at":"2026-05-28T01:25:37Z","abstract_excerpt":"Self-evolving agents improve over time by reflecting on past failures, but existing evaluation is limited in two ways: it measures only task scores, leaving reflection quality unknown, and it relies on agents' own episode runs, offering no mechanism to target specific failure patterns. We present \\textbf{BenchTrace}, a benchmark for evaluating self-evolution ability in LLM agents. BenchTrace is built on a snapshot-reflection dataset of 1,821 annotated episodes spanning six diverse tasks, and comprises a \\textbf{Reflection Evaluation} that probes failure identification through targeted QA tasks"},"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":"2605.29225","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-28T01:25:37Z","cross_cats_sorted":[],"title_canon_sha256":"5633ea62263f21d20d87b53696df8c11ec6d3c6bdbf8dd4a2df428bd7a237352","abstract_canon_sha256":"79a474ec236b8c5bfafda4c1f97072ada55fbb100eac3b158e14763616023f7c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:05:25.707714Z","signature_b64":"LpHU8axRzbLflXxF9VQ7RI58PNC5w+cTzZNJ7w7cr9YaEALN3h1BPb1titlOkcqqI1YVK0Hw6gyOB0lTPiiFDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2f08671d662223751fe3d864d268a59197451741bb0d370e94a5a925f2451812","last_reissued_at":"2026-05-29T01:05:25.706893Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:05:25.706893Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BenchTrace: A Benchmark for Testing Reflection Ability and Controlled Evolution in LLM Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Akiko Aizawa, Fei Cheng, Jiahao Huang, Junfeng Jiang, Zefan Yu","submitted_at":"2026-05-28T01:25:37Z","abstract_excerpt":"Self-evolving agents improve over time by reflecting on past failures, but existing evaluation is limited in two ways: it measures only task scores, leaving reflection quality unknown, and it relies on agents' own episode runs, offering no mechanism to target specific failure patterns. We present \\textbf{BenchTrace}, a benchmark for evaluating self-evolution ability in LLM agents. BenchTrace is built on a snapshot-reflection dataset of 1,821 annotated episodes spanning six diverse tasks, and comprises a \\textbf{Reflection Evaluation} that probes failure identification through targeted QA tasks"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.29225","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/2605.29225/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":"2605.29225","created_at":"2026-05-29T01:05:25.707024+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.29225v1","created_at":"2026-05-29T01:05:25.707024+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.29225","created_at":"2026-05-29T01:05:25.707024+00:00"},{"alias_kind":"pith_short_12","alias_value":"F4EGOHLGEIRX","created_at":"2026-05-29T01:05:25.707024+00:00"},{"alias_kind":"pith_short_16","alias_value":"F4EGOHLGEIRXKH7D","created_at":"2026-05-29T01:05:25.707024+00:00"},{"alias_kind":"pith_short_8","alias_value":"F4EGOHLG","created_at":"2026-05-29T01:05:25.707024+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":11,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/F4EGOHLGEIRXKH7D3BSNE2FFSG","json":"https://pith.science/pith/F4EGOHLGEIRXKH7D3BSNE2FFSG.json","graph_json":"https://pith.science/api/pith-number/F4EGOHLGEIRXKH7D3BSNE2FFSG/graph.json","events_json":"https://pith.science/api/pith-number/F4EGOHLGEIRXKH7D3BSNE2FFSG/events.json","paper":"https://pith.science/paper/F4EGOHLG"},"agent_actions":{"view_html":"https://pith.science/pith/F4EGOHLGEIRXKH7D3BSNE2FFSG","download_json":"https://pith.science/pith/F4EGOHLGEIRXKH7D3BSNE2FFSG.json","view_paper":"https://pith.science/paper/F4EGOHLG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.29225&json=true","fetch_graph":"https://pith.science/api/pith-number/F4EGOHLGEIRXKH7D3BSNE2FFSG/graph.json","fetch_events":"https://pith.science/api/pith-number/F4EGOHLGEIRXKH7D3BSNE2FFSG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F4EGOHLGEIRXKH7D3BSNE2FFSG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F4EGOHLGEIRXKH7D3BSNE2FFSG/action/storage_attestation","attest_author":"https://pith.science/pith/F4EGOHLGEIRXKH7D3BSNE2FFSG/action/author_attestation","sign_citation":"https://pith.science/pith/F4EGOHLGEIRXKH7D3BSNE2FFSG/action/citation_signature","submit_replication":"https://pith.science/pith/F4EGOHLGEIRXKH7D3BSNE2FFSG/action/replication_record"}},"created_at":"2026-05-29T01:05:25.707024+00:00","updated_at":"2026-05-29T01:05:25.707024+00:00"}