{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:C2XV2VPJWNE6BLCBPCMEVUIQ55","short_pith_number":"pith:C2XV2VPJ","schema_version":"1.0","canonical_sha256":"16af5d55e9b349e0ac4178984ad110ef44511a9b6b48b48251a19d3d21473406","source":{"kind":"arxiv","id":"2506.16440","version":1},"attestation_state":"computed","paper":{"title":"Evaluating the Use of LLMs for Documentation to Code Traceability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.SE","authors_text":"Ebube Alor, Emad Shihab, SayedHassan Khatoonabadi","submitted_at":"2025-06-19T16:18:53Z","abstract_excerpt":"Large Language Models (LLMs) offer new potential for automating documentation-to-code traceability, yet their capabilities remain underexplored. We present a comprehensive evaluation of LLMs (Claude 3.5 Sonnet, GPT-4o, and o3-mini) in establishing trace links between various software documentation (including API references and user guides) and source code. We create two novel datasets from two open-source projects (Unity Catalog and Crawl4AI). Through systematic experiments, we assess three key capabilities: (1) trace link identification accuracy, (2) relationship explanation quality, and (3) "},"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":"2506.16440","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2025-06-19T16:18:53Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"badb8fa95df3cd795969c85ecebf5305530fd35140b3f29de1d1ba56d5c5bf6c","abstract_canon_sha256":"b55a0ad74c111305edf9143bc6b6708f19f7f7e8e95d83ef8cf116ed21d72b6c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:49:55.852941Z","signature_b64":"kU7T+DrtiBLVMK7Nz0aB3RmAZrjQRUYGyRvgYFKnnK6Pv+ubaCVSj50R/c1jcKMV8NWPDzYOFzmZZrYpprNcCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"16af5d55e9b349e0ac4178984ad110ef44511a9b6b48b48251a19d3d21473406","last_reissued_at":"2026-07-05T11:49:55.852487Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:49:55.852487Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Evaluating the Use of LLMs for Documentation to Code Traceability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.SE","authors_text":"Ebube Alor, Emad Shihab, SayedHassan Khatoonabadi","submitted_at":"2025-06-19T16:18:53Z","abstract_excerpt":"Large Language Models (LLMs) offer new potential for automating documentation-to-code traceability, yet their capabilities remain underexplored. We present a comprehensive evaluation of LLMs (Claude 3.5 Sonnet, GPT-4o, and o3-mini) in establishing trace links between various software documentation (including API references and user guides) and source code. We create two novel datasets from two open-source projects (Unity Catalog and Crawl4AI). Through systematic experiments, we assess three key capabilities: (1) trace link identification accuracy, (2) relationship explanation quality, and (3) "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.16440","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/2506.16440/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":"2506.16440","created_at":"2026-07-05T11:49:55.852543+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.16440v1","created_at":"2026-07-05T11:49:55.852543+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.16440","created_at":"2026-07-05T11:49:55.852543+00:00"},{"alias_kind":"pith_short_12","alias_value":"C2XV2VPJWNE6","created_at":"2026-07-05T11:49:55.852543+00:00"},{"alias_kind":"pith_short_16","alias_value":"C2XV2VPJWNE6BLCB","created_at":"2026-07-05T11:49:55.852543+00:00"},{"alias_kind":"pith_short_8","alias_value":"C2XV2VPJ","created_at":"2026-07-05T11:49:55.852543+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24834","citing_title":"Accuracy and Satisfaction in Multi-Turn LLM Dialogues for NFR Assessment","ref_index":95,"is_internal_anchor":false},{"citing_arxiv_id":"2604.22432","citing_title":"R2Code: A Self-Reflective LLM Framework for Requirements-to-Code Traceability","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/C2XV2VPJWNE6BLCBPCMEVUIQ55","json":"https://pith.science/pith/C2XV2VPJWNE6BLCBPCMEVUIQ55.json","graph_json":"https://pith.science/api/pith-number/C2XV2VPJWNE6BLCBPCMEVUIQ55/graph.json","events_json":"https://pith.science/api/pith-number/C2XV2VPJWNE6BLCBPCMEVUIQ55/events.json","paper":"https://pith.science/paper/C2XV2VPJ"},"agent_actions":{"view_html":"https://pith.science/pith/C2XV2VPJWNE6BLCBPCMEVUIQ55","download_json":"https://pith.science/pith/C2XV2VPJWNE6BLCBPCMEVUIQ55.json","view_paper":"https://pith.science/paper/C2XV2VPJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.16440&json=true","fetch_graph":"https://pith.science/api/pith-number/C2XV2VPJWNE6BLCBPCMEVUIQ55/graph.json","fetch_events":"https://pith.science/api/pith-number/C2XV2VPJWNE6BLCBPCMEVUIQ55/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/C2XV2VPJWNE6BLCBPCMEVUIQ55/action/timestamp_anchor","attest_storage":"https://pith.science/pith/C2XV2VPJWNE6BLCBPCMEVUIQ55/action/storage_attestation","attest_author":"https://pith.science/pith/C2XV2VPJWNE6BLCBPCMEVUIQ55/action/author_attestation","sign_citation":"https://pith.science/pith/C2XV2VPJWNE6BLCBPCMEVUIQ55/action/citation_signature","submit_replication":"https://pith.science/pith/C2XV2VPJWNE6BLCBPCMEVUIQ55/action/replication_record"}},"created_at":"2026-07-05T11:49:55.852543+00:00","updated_at":"2026-07-05T11:49:55.852543+00:00"}