{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:CV3CA3QTTSQJLC4NWMLPMUZJ4W","short_pith_number":"pith:CV3CA3QT","schema_version":"1.0","canonical_sha256":"1576206e139ca0958b8db316f65329e5bbc83057e83fc534c5b8b39ff362a24e","source":{"kind":"arxiv","id":"2503.03149","version":1},"attestation_state":"computed","paper":{"title":"DSVD: Dynamic Self-Verify Decoding for Faithful Generation in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Pingjie Wang, Yanfeng Wang, Ya Zhang, Yiqiu Guo, Yuchen Yang, Yusheng Liao, Yu Wang, Zhe Chen","submitted_at":"2025-03-05T03:45:50Z","abstract_excerpt":"The reliability of large language models remains a critical challenge, particularly due to their susceptibility to hallucinations and factual inaccuracies during text generation. Existing solutions either underutilize models' self-correction with preemptive strategies or use costly post-hoc verification. To further explore the potential of real-time self-verification and correction, we present Dynamic Self-Verify Decoding (DSVD), a novel decoding framework that enhances generation reliability through real-time hallucination detection and efficient error correction. DSVD integrates two key comp"},"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":"2503.03149","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-05T03:45:50Z","cross_cats_sorted":[],"title_canon_sha256":"36731774c3c5ae7d4cd154358a90e24e8874be5a21ab236d970732787767d613","abstract_canon_sha256":"6e35b4f389b15fac9723f00f8e72a634578368a99d10effc27bb81fb531a52ce"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:24:40.168998Z","signature_b64":"KggSp52z+Gp1Y791oPa1Py+waArtS23tggUwm7j+lvAemPjn5WgfQ0+WsbIdPfHK/M6iQkDpv/+4V+jc66+4Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1576206e139ca0958b8db316f65329e5bbc83057e83fc534c5b8b39ff362a24e","last_reissued_at":"2026-07-05T10:24:40.168472Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:24:40.168472Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DSVD: Dynamic Self-Verify Decoding for Faithful Generation in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Pingjie Wang, Yanfeng Wang, Ya Zhang, Yiqiu Guo, Yuchen Yang, Yusheng Liao, Yu Wang, Zhe Chen","submitted_at":"2025-03-05T03:45:50Z","abstract_excerpt":"The reliability of large language models remains a critical challenge, particularly due to their susceptibility to hallucinations and factual inaccuracies during text generation. Existing solutions either underutilize models' self-correction with preemptive strategies or use costly post-hoc verification. To further explore the potential of real-time self-verification and correction, we present Dynamic Self-Verify Decoding (DSVD), a novel decoding framework that enhances generation reliability through real-time hallucination detection and efficient error correction. DSVD integrates two key comp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.03149","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/2503.03149/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":"2503.03149","created_at":"2026-07-05T10:24:40.168550+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.03149v1","created_at":"2026-07-05T10:24:40.168550+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.03149","created_at":"2026-07-05T10:24:40.168550+00:00"},{"alias_kind":"pith_short_12","alias_value":"CV3CA3QTTSQJ","created_at":"2026-07-05T10:24:40.168550+00:00"},{"alias_kind":"pith_short_16","alias_value":"CV3CA3QTTSQJLC4N","created_at":"2026-07-05T10:24:40.168550+00:00"},{"alias_kind":"pith_short_8","alias_value":"CV3CA3QT","created_at":"2026-07-05T10:24:40.168550+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.02170","citing_title":"Avoidance Decoding for Diverse Multi-Branch Story Generation","ref_index":10,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CV3CA3QTTSQJLC4NWMLPMUZJ4W","json":"https://pith.science/pith/CV3CA3QTTSQJLC4NWMLPMUZJ4W.json","graph_json":"https://pith.science/api/pith-number/CV3CA3QTTSQJLC4NWMLPMUZJ4W/graph.json","events_json":"https://pith.science/api/pith-number/CV3CA3QTTSQJLC4NWMLPMUZJ4W/events.json","paper":"https://pith.science/paper/CV3CA3QT"},"agent_actions":{"view_html":"https://pith.science/pith/CV3CA3QTTSQJLC4NWMLPMUZJ4W","download_json":"https://pith.science/pith/CV3CA3QTTSQJLC4NWMLPMUZJ4W.json","view_paper":"https://pith.science/paper/CV3CA3QT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.03149&json=true","fetch_graph":"https://pith.science/api/pith-number/CV3CA3QTTSQJLC4NWMLPMUZJ4W/graph.json","fetch_events":"https://pith.science/api/pith-number/CV3CA3QTTSQJLC4NWMLPMUZJ4W/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CV3CA3QTTSQJLC4NWMLPMUZJ4W/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CV3CA3QTTSQJLC4NWMLPMUZJ4W/action/storage_attestation","attest_author":"https://pith.science/pith/CV3CA3QTTSQJLC4NWMLPMUZJ4W/action/author_attestation","sign_citation":"https://pith.science/pith/CV3CA3QTTSQJLC4NWMLPMUZJ4W/action/citation_signature","submit_replication":"https://pith.science/pith/CV3CA3QTTSQJLC4NWMLPMUZJ4W/action/replication_record"}},"created_at":"2026-07-05T10:24:40.168550+00:00","updated_at":"2026-07-05T10:24:40.168550+00:00"}