{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YZAAUP3EMHD2EPNG3655IVTYBL","short_pith_number":"pith:YZAAUP3E","schema_version":"1.0","canonical_sha256":"c6400a3f6461c7a23da6dfbbd456780ac4a756066a136476fca236ce5f73df9d","source":{"kind":"arxiv","id":"2509.04499","version":1},"attestation_state":"computed","paper":{"title":"DeepTRACE: Auditing Deep Research AI Systems for Tracking Reliability Across Citations and Evidence","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chien-Sheng Wu, Kung-Hsiang Huang, Philippe Laban, Pranav Narayanan Venkit, Yilun Zhou, Yixin Mao","submitted_at":"2025-09-02T00:32:38Z","abstract_excerpt":"Generative search engines and deep research LLM agents promise trustworthy, source-grounded synthesis, yet users regularly encounter overconfidence, weak sourcing, and confusing citation practices. We introduce DeepTRACE, a novel sociotechnically grounded audit framework that turns prior community-identified failure cases into eight measurable dimensions spanning answer text, sources, and citations. DeepTRACE uses statement-level analysis (decomposition, confidence scoring) and builds citation and factual-support matrices to audit how systems reason with and attribute evidence end-to-end. Usin"},"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":"2509.04499","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-09-02T00:32:38Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"55544485e0f1ed6971d10cc981d79594f9a773b1e73001ba785d9ef9adc890e4","abstract_canon_sha256":"77b9a6ba477feb2a9d27f6d764a73d19fb8f613230a7cf41fd0b6bfcd4a9a966"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:05:24.035363Z","signature_b64":"dpyYC5/cKBkIhVzPlzvt1j2xHQ+6ebBoLSyhWwuxcCMQEki/iEGam4BoD3d/4tsb908wDcJFhOgY2IecAVRkBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c6400a3f6461c7a23da6dfbbd456780ac4a756066a136476fca236ce5f73df9d","last_reissued_at":"2026-07-05T12:05:24.034893Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:05:24.034893Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DeepTRACE: Auditing Deep Research AI Systems for Tracking Reliability Across Citations and Evidence","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chien-Sheng Wu, Kung-Hsiang Huang, Philippe Laban, Pranav Narayanan Venkit, Yilun Zhou, Yixin Mao","submitted_at":"2025-09-02T00:32:38Z","abstract_excerpt":"Generative search engines and deep research LLM agents promise trustworthy, source-grounded synthesis, yet users regularly encounter overconfidence, weak sourcing, and confusing citation practices. We introduce DeepTRACE, a novel sociotechnically grounded audit framework that turns prior community-identified failure cases into eight measurable dimensions spanning answer text, sources, and citations. DeepTRACE uses statement-level analysis (decomposition, confidence scoring) and builds citation and factual-support matrices to audit how systems reason with and attribute evidence end-to-end. Usin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.04499","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/2509.04499/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":"2509.04499","created_at":"2026-07-05T12:05:24.034948+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.04499v1","created_at":"2026-07-05T12:05:24.034948+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.04499","created_at":"2026-07-05T12:05:24.034948+00:00"},{"alias_kind":"pith_short_12","alias_value":"YZAAUP3EMHD2","created_at":"2026-07-05T12:05:24.034948+00:00"},{"alias_kind":"pith_short_16","alias_value":"YZAAUP3EMHD2EPNG","created_at":"2026-07-05T12:05:24.034948+00:00"},{"alias_kind":"pith_short_8","alias_value":"YZAAUP3E","created_at":"2026-07-05T12:05:24.034948+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.07896","citing_title":"What if AI systems weren't chatbots?","ref_index":185,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YZAAUP3EMHD2EPNG3655IVTYBL","json":"https://pith.science/pith/YZAAUP3EMHD2EPNG3655IVTYBL.json","graph_json":"https://pith.science/api/pith-number/YZAAUP3EMHD2EPNG3655IVTYBL/graph.json","events_json":"https://pith.science/api/pith-number/YZAAUP3EMHD2EPNG3655IVTYBL/events.json","paper":"https://pith.science/paper/YZAAUP3E"},"agent_actions":{"view_html":"https://pith.science/pith/YZAAUP3EMHD2EPNG3655IVTYBL","download_json":"https://pith.science/pith/YZAAUP3EMHD2EPNG3655IVTYBL.json","view_paper":"https://pith.science/paper/YZAAUP3E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.04499&json=true","fetch_graph":"https://pith.science/api/pith-number/YZAAUP3EMHD2EPNG3655IVTYBL/graph.json","fetch_events":"https://pith.science/api/pith-number/YZAAUP3EMHD2EPNG3655IVTYBL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YZAAUP3EMHD2EPNG3655IVTYBL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YZAAUP3EMHD2EPNG3655IVTYBL/action/storage_attestation","attest_author":"https://pith.science/pith/YZAAUP3EMHD2EPNG3655IVTYBL/action/author_attestation","sign_citation":"https://pith.science/pith/YZAAUP3EMHD2EPNG3655IVTYBL/action/citation_signature","submit_replication":"https://pith.science/pith/YZAAUP3EMHD2EPNG3655IVTYBL/action/replication_record"}},"created_at":"2026-07-05T12:05:24.034948+00:00","updated_at":"2026-07-05T12:05:24.034948+00:00"}