{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:2XD7AD2DV4LE6DKOARAWLZULHA","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":"803d155a77ded41b300fe34dced984f3eb46a0af65c306e31432ddf0c8f2c16a","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-05-13T03:00:57Z","title_canon_sha256":"07872682c0cfb5124101ff197f31db38f83c886e616dae8057e2bc39abffda39"},"schema_version":"1.0","source":{"id":"2605.12925","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.12925","created_at":"2026-05-18T03:09:10Z"},{"alias_kind":"arxiv_version","alias_value":"2605.12925v1","created_at":"2026-05-18T03:09:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.12925","created_at":"2026-05-18T03:09:10Z"},{"alias_kind":"pith_short_12","alias_value":"2XD7AD2DV4LE","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"2XD7AD2DV4LE6DKO","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"2XD7AD2D","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:8340795877791325c5d9ab593c559490d591b61f08eecc131ca6693052139e2b","target":"graph","created_at":"2026-05-18T03:09:10Z","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":"Among passing trajectories in this subset, 10.7% exhibit behavior we call a Lucky Pass: regression cycles, blind retries, missing verification, or temporally disordered exploration, implementation, and verification."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That Prefix Tree Acceptor references built by merging multiple passing solutions accurately represent principled behavior without incorporating lucky elements from the source trajectories."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"10.7% of passing SWE-agent trajectories are Lucky Passes with chaotic behaviors, and a quality score based on process references changes model rankings across eight backends."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Binary pass rates in SWE-agent tests equate chaotic trial-and-error successes with systematic ones."}],"snapshot_sha256":"adce0b450020fee18eaa30704ab28f998777265f989745261d06c17949418e61"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Evaluation of software engineering (SWE) agents is dominated by a binary signal: whether the final patch passes the tests. This outcome-only view treats a principled solution and a chaotic trial-and-error process as equivalent. We show that this equivalence is empirically false. We evaluate 2,614 OpenHands trajectories from eight model backends on 60 SWE-bench Verified tasks. Of these, 47 have enough passing trajectories to construct task-level process references, yielding a 1,815-trajectory evaluation subset. Among passing trajectories in this subset, 10.7% exhibit behavior we call a Lucky Pa","authors_text":"Benjamin Steenhoek, Gaurav Mittal, Pingping Lin, Priyam Sahoo, Shengjie Ma, Xiaomin Li, Yu Hu","cross_cats":["cs.AI"],"headline":"Binary pass rates in SWE-agent tests equate chaotic trial-and-error successes with systematic ones.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-05-13T03:00:57Z","title":"AgentLens: Revealing The Lucky Pass Problem in SWE-Agent Evaluation"},"references":{"count":61,"internal_anchors":14,"resolved_work":61,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"arXiv preprint arXiv:2410.20285 , year=","work_id":"e26e6ec9-37e3-47d8-b494-e0a227e77e36","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Swe-rebench: An automated pipeline for task collection and decontaminated evaluation of software engineering agents","work_id":"c819f5fe-32b6-41d1-aeb5-2d80c6fa8474","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Islem Bouzenia, Premkumar Devanbu, and Michael Pradel","work_id":"9584953a-2ac8-49b2-bcd3-44924077f86d","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"2026 , note =","work_id":"ad7bad76-b98f-4522-8476-a382ae24a4b8","year":2026},{"cited_arxiv_id":"","doi":"10.5281/zenodo.19357078","is_internal_anchor":false,"ref_index":5,"title":"doi:10.5281/zenodo.19357078 , url =","work_id":"5aab64ed-4762-4aa1-922b-18ae4b5b8b27","year":null}],"snapshot_sha256":"66df3e8e7819f1fb70a7bba3a29167044596023591235646b3882339a9f38ab8"},"source":{"id":"2605.12925","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-14T18:46:32.801828Z","id":"65a2b27d-f897-4cc8-8786-4ee98826b4a1","model_set":{"reader":"grok-4.3"},"one_line_summary":"10.7% of passing SWE-agent trajectories are Lucky Passes with chaotic behaviors, and a quality score based on process references changes model rankings across eight backends.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Binary pass rates in SWE-agent tests equate chaotic trial-and-error successes with systematic ones.","strongest_claim":"Among passing trajectories in this subset, 10.7% exhibit behavior we call a Lucky Pass: regression cycles, blind retries, missing verification, or temporally disordered exploration, implementation, and verification.","weakest_assumption":"That Prefix Tree Acceptor references built by merging multiple passing solutions accurately represent principled behavior without incorporating lucky elements from the source trajectories."}},"verdict_id":"65a2b27d-f897-4cc8-8786-4ee98826b4a1"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:7710c5082c6465967c4e94b3e6a845859df7f52d75a06020cc3b3e886279ca57","target":"record","created_at":"2026-05-18T03:09:10Z","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":"803d155a77ded41b300fe34dced984f3eb46a0af65c306e31432ddf0c8f2c16a","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-05-13T03:00:57Z","title_canon_sha256":"07872682c0cfb5124101ff197f31db38f83c886e616dae8057e2bc39abffda39"},"schema_version":"1.0","source":{"id":"2605.12925","kind":"arxiv","version":1}},"canonical_sha256":"d5c7f00f43af164f0d4e044165e68b380302144e136f8c90ba9bcc3473f5877d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d5c7f00f43af164f0d4e044165e68b380302144e136f8c90ba9bcc3473f5877d","first_computed_at":"2026-05-18T03:09:10.100969Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T03:09:10.100969Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"t5jAyHzJ6WH8ts/9WL17z+PDjmmPs7j5Ynp59GedCrqQnODuZVxSDQiFYZxm5mtpRIo9y51ZUIcopL6M1bN9CA==","signature_status":"signed_v1","signed_at":"2026-05-18T03:09:10.101635Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.12925","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7710c5082c6465967c4e94b3e6a845859df7f52d75a06020cc3b3e886279ca57","sha256:8340795877791325c5d9ab593c559490d591b61f08eecc131ca6693052139e2b"],"state_sha256":"ffdec1d8fe338ecd3147807d19ae3e3d02a6ba89e4823e93865d2f3f2aac3d9c"}