{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ACSAH2A55CFKXCOHCHJ3BKI6DT","short_pith_number":"pith:ACSAH2A5","schema_version":"1.0","canonical_sha256":"00a403e81de88aab89c711d3b0a91e1ce66a055a9c72e288016d5808db6815c6","source":{"kind":"arxiv","id":"2406.13692","version":2},"attestation_state":"computed","paper":{"title":"Synchronous Faithfulness Monitoring for Trustworthy Retrieval-Augmented Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Di Wu, Fan Yin, Jia-Chen Gu, Kai-Wei Chang, Nanyun Peng","submitted_at":"2024-06-19T16:42:57Z","abstract_excerpt":"Retrieval-augmented language models (RALMs) have shown strong performance and wide applicability in knowledge-intensive tasks. However, there are significant trustworthiness concerns as RALMs are prone to generating unfaithful outputs, including baseless information or contradictions with the retrieved context. This paper proposes SynCheck, a lightweight monitor that leverages fine-grained decoding dynamics including sequence likelihood, uncertainty quantification, context influence, and semantic alignment to synchronously detect unfaithful sentences. By integrating efficiently measurable and "},"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":"2406.13692","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-19T16:42:57Z","cross_cats_sorted":[],"title_canon_sha256":"e0ada4f3fb8be5d66767b7f9df5c2f4305be923a10894aa2c6ab471ae085709b","abstract_canon_sha256":"e525b0f5a4cdd2686e5d48e4a34839f75bf33875d42ccf92385345c8c7f21fd9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:15:34.070104Z","signature_b64":"E2OjAu8DKP61XxM7Dk9cVBut3FBWQ2/evz0mHRzC8HCUWGcoCqqRBXRiGsOZUGEIOpVyA9zT/cFlHpa8k4uqAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"00a403e81de88aab89c711d3b0a91e1ce66a055a9c72e288016d5808db6815c6","last_reissued_at":"2026-07-05T09:15:34.069600Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:15:34.069600Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Synchronous Faithfulness Monitoring for Trustworthy Retrieval-Augmented Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Di Wu, Fan Yin, Jia-Chen Gu, Kai-Wei Chang, Nanyun Peng","submitted_at":"2024-06-19T16:42:57Z","abstract_excerpt":"Retrieval-augmented language models (RALMs) have shown strong performance and wide applicability in knowledge-intensive tasks. However, there are significant trustworthiness concerns as RALMs are prone to generating unfaithful outputs, including baseless information or contradictions with the retrieved context. This paper proposes SynCheck, a lightweight monitor that leverages fine-grained decoding dynamics including sequence likelihood, uncertainty quantification, context influence, and semantic alignment to synchronously detect unfaithful sentences. By integrating efficiently measurable and "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.13692","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/2406.13692/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":"2406.13692","created_at":"2026-07-05T09:15:34.069665+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.13692v2","created_at":"2026-07-05T09:15:34.069665+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.13692","created_at":"2026-07-05T09:15:34.069665+00:00"},{"alias_kind":"pith_short_12","alias_value":"ACSAH2A55CFK","created_at":"2026-07-05T09:15:34.069665+00:00"},{"alias_kind":"pith_short_16","alias_value":"ACSAH2A55CFKXCOH","created_at":"2026-07-05T09:15:34.069665+00:00"},{"alias_kind":"pith_short_8","alias_value":"ACSAH2A5","created_at":"2026-07-05T09:15:34.069665+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.00646","citing_title":"RAG-PRISM: A Personalized, Rapid, and Immersive Skill Mastery Framework with Adaptive Retrieval-Augmented Tutoring","ref_index":33,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ACSAH2A55CFKXCOHCHJ3BKI6DT","json":"https://pith.science/pith/ACSAH2A55CFKXCOHCHJ3BKI6DT.json","graph_json":"https://pith.science/api/pith-number/ACSAH2A55CFKXCOHCHJ3BKI6DT/graph.json","events_json":"https://pith.science/api/pith-number/ACSAH2A55CFKXCOHCHJ3BKI6DT/events.json","paper":"https://pith.science/paper/ACSAH2A5"},"agent_actions":{"view_html":"https://pith.science/pith/ACSAH2A55CFKXCOHCHJ3BKI6DT","download_json":"https://pith.science/pith/ACSAH2A55CFKXCOHCHJ3BKI6DT.json","view_paper":"https://pith.science/paper/ACSAH2A5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.13692&json=true","fetch_graph":"https://pith.science/api/pith-number/ACSAH2A55CFKXCOHCHJ3BKI6DT/graph.json","fetch_events":"https://pith.science/api/pith-number/ACSAH2A55CFKXCOHCHJ3BKI6DT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ACSAH2A55CFKXCOHCHJ3BKI6DT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ACSAH2A55CFKXCOHCHJ3BKI6DT/action/storage_attestation","attest_author":"https://pith.science/pith/ACSAH2A55CFKXCOHCHJ3BKI6DT/action/author_attestation","sign_citation":"https://pith.science/pith/ACSAH2A55CFKXCOHCHJ3BKI6DT/action/citation_signature","submit_replication":"https://pith.science/pith/ACSAH2A55CFKXCOHCHJ3BKI6DT/action/replication_record"}},"created_at":"2026-07-05T09:15:34.069665+00:00","updated_at":"2026-07-05T09:15:34.069665+00:00"}