{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:QVGMWXIJRB753WFIRLHJKIWVBA","short_pith_number":"pith:QVGMWXIJ","schema_version":"1.0","canonical_sha256":"854ccb5d09887fddd8a88ace9522d50815bc98b86714aa13ae398b97a28d5884","source":{"kind":"arxiv","id":"2409.00729","version":2},"attestation_state":"computed","paper":{"title":"ContextCite: Attributing Model Generation to Context","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Aleksander Madry, Benjamin Cohen-Wang, Harshay Shah, Kristian Georgiev","submitted_at":"2024-09-01T14:36:36Z","abstract_excerpt":"How do language models use information provided as context when generating a response? Can we infer whether a particular generated statement is actually grounded in the context, a misinterpretation, or fabricated? To help answer these questions, we introduce the problem of context attribution: pinpointing the parts of the context (if any) that led a model to generate a particular statement. We then present ContextCite, a simple and scalable method for context attribution that can be applied on top of any existing language model. Finally, we showcase the utility of ContextCite through three app"},"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":"2409.00729","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-09-01T14:36:36Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"eab51fe8b8338a0e789750df88063aab167a557ae8c7b9732c3735e897a31a9b","abstract_canon_sha256":"a926e2da5cb867e8d1ea931361314be2f03a5bfc5059413369d66e83327df009"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:07:02.779258Z","signature_b64":"vn44tWUTUkQvNzyxCHtOafkpwdtH5L0Dk2xWuIXaPBN6+7iUjSCd3gVB8Tfa4041QRyqxVgDFjhwgglNc78FBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"854ccb5d09887fddd8a88ace9522d50815bc98b86714aa13ae398b97a28d5884","last_reissued_at":"2026-07-05T09:07:02.778838Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:07:02.778838Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ContextCite: Attributing Model Generation to Context","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Aleksander Madry, Benjamin Cohen-Wang, Harshay Shah, Kristian Georgiev","submitted_at":"2024-09-01T14:36:36Z","abstract_excerpt":"How do language models use information provided as context when generating a response? Can we infer whether a particular generated statement is actually grounded in the context, a misinterpretation, or fabricated? To help answer these questions, we introduce the problem of context attribution: pinpointing the parts of the context (if any) that led a model to generate a particular statement. We then present ContextCite, a simple and scalable method for context attribution that can be applied on top of any existing language model. Finally, we showcase the utility of ContextCite through three app"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.00729","kind":"arxiv","version":2},"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/2409.00729/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":"2409.00729","created_at":"2026-07-05T09:07:02.778894+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.00729v2","created_at":"2026-07-05T09:07:02.778894+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.00729","created_at":"2026-07-05T09:07:02.778894+00:00"},{"alias_kind":"pith_short_12","alias_value":"QVGMWXIJRB75","created_at":"2026-07-05T09:07:02.778894+00:00"},{"alias_kind":"pith_short_16","alias_value":"QVGMWXIJRB753WFI","created_at":"2026-07-05T09:07:02.778894+00:00"},{"alias_kind":"pith_short_8","alias_value":"QVGMWXIJ","created_at":"2026-07-05T09:07:02.778894+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26449","citing_title":"ProvenAI: Provenance-Native Traces of Evidence in Generated Answers","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28358","citing_title":"How Do LLMs Cite? A Mechanistic Interpretation of Attribution in Retrieval-Augmented Generation","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.22237","citing_title":"Tell Me Why: Designing an Explainable LLM-based Dialogue System for Student Problem Behavior Diagnosis","ref_index":4,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QVGMWXIJRB753WFIRLHJKIWVBA","json":"https://pith.science/pith/QVGMWXIJRB753WFIRLHJKIWVBA.json","graph_json":"https://pith.science/api/pith-number/QVGMWXIJRB753WFIRLHJKIWVBA/graph.json","events_json":"https://pith.science/api/pith-number/QVGMWXIJRB753WFIRLHJKIWVBA/events.json","paper":"https://pith.science/paper/QVGMWXIJ"},"agent_actions":{"view_html":"https://pith.science/pith/QVGMWXIJRB753WFIRLHJKIWVBA","download_json":"https://pith.science/pith/QVGMWXIJRB753WFIRLHJKIWVBA.json","view_paper":"https://pith.science/paper/QVGMWXIJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.00729&json=true","fetch_graph":"https://pith.science/api/pith-number/QVGMWXIJRB753WFIRLHJKIWVBA/graph.json","fetch_events":"https://pith.science/api/pith-number/QVGMWXIJRB753WFIRLHJKIWVBA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QVGMWXIJRB753WFIRLHJKIWVBA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QVGMWXIJRB753WFIRLHJKIWVBA/action/storage_attestation","attest_author":"https://pith.science/pith/QVGMWXIJRB753WFIRLHJKIWVBA/action/author_attestation","sign_citation":"https://pith.science/pith/QVGMWXIJRB753WFIRLHJKIWVBA/action/citation_signature","submit_replication":"https://pith.science/pith/QVGMWXIJRB753WFIRLHJKIWVBA/action/replication_record"}},"created_at":"2026-07-05T09:07:02.778894+00:00","updated_at":"2026-07-05T09:07:02.778894+00:00"}