{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:FTX3HTHEAHSFUNCCEDNEY7CPLG","short_pith_number":"pith:FTX3HTHE","schema_version":"1.0","canonical_sha256":"2cefb3cce401e45a344220da4c7c4f59a716a749dd2f65fd89365e59abd23e2d","source":{"kind":"arxiv","id":"2605.18332","version":1},"attestation_state":"computed","paper":{"title":"Same Signal, Different Semantics: A Cross-Framework Behavioral Analysis of Software Engineering Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Jingxu Gu, Lingxiao Jiang, Shangqing Liu, Tianling Li, Wei Ma, Zhi Chen","submitted_at":"2026-05-18T12:49:18Z","abstract_excerpt":"Behavioral studies of LLM-based software engineering agents extract operational rules about which trajectory shapes correlate with higher resolution rates: that a test step follows a code modification, that error cascades are short, or that trajectories are compact. Each rule is typically derived from a single framework, and whether it transfers, in sign as well as magnitude, to structurally different agent designs has not been directly tested.\n  We address this at ecosystem scale: 64,380 SWE-bench runs from 126 agent configurations spanning 43 frameworks, where each configuration pairs an LLM"},"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":"2605.18332","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2026-05-18T12:49:18Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d293bab2453e59e6c27cb8a3991ed5c476fdb897d506f06cb018c50e562bf610","abstract_canon_sha256":"3d38b594681e6e8373c83b68846bfc75fffe0c294167daea8181a9783f230278"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:05:55.649845Z","signature_b64":"DePI3ruSTzkSFva7KXOD55nveYJn1/Lq+UcC6jFM9/dOl5kT8710stPhVy9lGE5D1ROxYY5kOpaaPQ+ddrX+Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2cefb3cce401e45a344220da4c7c4f59a716a749dd2f65fd89365e59abd23e2d","last_reissued_at":"2026-05-20T00:05:55.648951Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:05:55.648951Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Same Signal, Different Semantics: A Cross-Framework Behavioral Analysis of Software Engineering Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Jingxu Gu, Lingxiao Jiang, Shangqing Liu, Tianling Li, Wei Ma, Zhi Chen","submitted_at":"2026-05-18T12:49:18Z","abstract_excerpt":"Behavioral studies of LLM-based software engineering agents extract operational rules about which trajectory shapes correlate with higher resolution rates: that a test step follows a code modification, that error cascades are short, or that trajectories are compact. Each rule is typically derived from a single framework, and whether it transfers, in sign as well as magnitude, to structurally different agent designs has not been directly tested.\n  We address this at ecosystem scale: 64,380 SWE-bench runs from 126 agent configurations spanning 43 frameworks, where each configuration pairs an LLM"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.18332","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/2605.18332/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-19T23:33:35.174450Z","status":"skipped","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T23:21:58.843558Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"d1b9c87c70def7cec4b59437cd511b7605f6dd489d82060ea49403fca758f949"},"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":"2605.18332","created_at":"2026-05-20T00:05:55.649102+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.18332v1","created_at":"2026-05-20T00:05:55.649102+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.18332","created_at":"2026-05-20T00:05:55.649102+00:00"},{"alias_kind":"pith_short_12","alias_value":"FTX3HTHEAHSF","created_at":"2026-05-20T00:05:55.649102+00:00"},{"alias_kind":"pith_short_16","alias_value":"FTX3HTHEAHSFUNCC","created_at":"2026-05-20T00:05:55.649102+00:00"},{"alias_kind":"pith_short_8","alias_value":"FTX3HTHE","created_at":"2026-05-20T00:05:55.649102+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FTX3HTHEAHSFUNCCEDNEY7CPLG","json":"https://pith.science/pith/FTX3HTHEAHSFUNCCEDNEY7CPLG.json","graph_json":"https://pith.science/api/pith-number/FTX3HTHEAHSFUNCCEDNEY7CPLG/graph.json","events_json":"https://pith.science/api/pith-number/FTX3HTHEAHSFUNCCEDNEY7CPLG/events.json","paper":"https://pith.science/paper/FTX3HTHE"},"agent_actions":{"view_html":"https://pith.science/pith/FTX3HTHEAHSFUNCCEDNEY7CPLG","download_json":"https://pith.science/pith/FTX3HTHEAHSFUNCCEDNEY7CPLG.json","view_paper":"https://pith.science/paper/FTX3HTHE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.18332&json=true","fetch_graph":"https://pith.science/api/pith-number/FTX3HTHEAHSFUNCCEDNEY7CPLG/graph.json","fetch_events":"https://pith.science/api/pith-number/FTX3HTHEAHSFUNCCEDNEY7CPLG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FTX3HTHEAHSFUNCCEDNEY7CPLG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FTX3HTHEAHSFUNCCEDNEY7CPLG/action/storage_attestation","attest_author":"https://pith.science/pith/FTX3HTHEAHSFUNCCEDNEY7CPLG/action/author_attestation","sign_citation":"https://pith.science/pith/FTX3HTHEAHSFUNCCEDNEY7CPLG/action/citation_signature","submit_replication":"https://pith.science/pith/FTX3HTHEAHSFUNCCEDNEY7CPLG/action/replication_record"}},"created_at":"2026-05-20T00:05:55.649102+00:00","updated_at":"2026-05-20T00:05:55.649102+00:00"}