{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:NWPPO5TYZFU4WK2LCOM2UJYUBI","short_pith_number":"pith:NWPPO5TY","canonical_record":{"source":{"id":"2402.12545","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-19T21:12:14Z","cross_cats_sorted":[],"title_canon_sha256":"b52d139339ca25081f1d97645f67949d4945ce3138bde84294598d0fd9237745","abstract_canon_sha256":"1726be6dc8e15628de56f4cbe6a98462a7d4c4728f90dc9ed1cf99a241f987da"},"schema_version":"1.0"},"canonical_sha256":"6d9ef77678c969cb2b4b1399aa27140a293ecfdb764e85dc409ba30541809d64","source":{"kind":"arxiv","id":"2402.12545","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.12545","created_at":"2026-07-05T08:16:08Z"},{"alias_kind":"arxiv_version","alias_value":"2402.12545v2","created_at":"2026-07-05T08:16:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.12545","created_at":"2026-07-05T08:16:08Z"},{"alias_kind":"pith_short_12","alias_value":"NWPPO5TYZFU4","created_at":"2026-07-05T08:16:08Z"},{"alias_kind":"pith_short_16","alias_value":"NWPPO5TYZFU4WK2L","created_at":"2026-07-05T08:16:08Z"},{"alias_kind":"pith_short_8","alias_value":"NWPPO5TY","created_at":"2026-07-05T08:16:08Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:NWPPO5TYZFU4WK2LCOM2UJYUBI","target":"record","payload":{"canonical_record":{"source":{"id":"2402.12545","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-19T21:12:14Z","cross_cats_sorted":[],"title_canon_sha256":"b52d139339ca25081f1d97645f67949d4945ce3138bde84294598d0fd9237745","abstract_canon_sha256":"1726be6dc8e15628de56f4cbe6a98462a7d4c4728f90dc9ed1cf99a241f987da"},"schema_version":"1.0"},"canonical_sha256":"6d9ef77678c969cb2b4b1399aa27140a293ecfdb764e85dc409ba30541809d64","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:16:08.637692Z","signature_b64":"fO8nbWnd8lm5L7I+ODs5FSbthhM52dDdDXKHPYjWpqFmuNtkTJBk8/soP7XbKPFIBLzLQzvbwe1Bk/YgIDTQAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6d9ef77678c969cb2b4b1399aa27140a293ecfdb764e85dc409ba30541809d64","last_reissued_at":"2026-07-05T08:16:08.637218Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:16:08.637218Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.12545","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:16:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PnW4Pkbk+mKhZfHNcg6CzZg2oE7BRNXQL7MneXpS48Wzc5zwBkqr2lDISDYGW5Q9evMWdtRuMMqP6ShVXNDTCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T21:01:50.221363Z"},"content_sha256":"a7988d792c9d36af1e91e9c0a12e089b0f2b715e9c590926172a00131ce9ff43","schema_version":"1.0","event_id":"sha256:a7988d792c9d36af1e91e9c0a12e089b0f2b715e9c590926172a00131ce9ff43"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:NWPPO5TYZFU4WK2LCOM2UJYUBI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"TrustScore: Reference-Free Evaluation of LLM Response Trustworthiness","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Danna Zheng, Danyang Liu, Jeff Z. Pan, Mirella Lapata","submitted_at":"2024-02-19T21:12:14Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated impressive capabilities across various domains, prompting a surge in their practical applications. However, concerns have arisen regarding the trustworthiness of LLMs outputs, particularly in closed-book question-answering tasks, where non-experts may struggle to identify inaccuracies due to the absence of contextual or ground truth information. This paper introduces TrustScore, a framework based on the concept of Behavioral Consistency, which evaluates whether an LLMs response aligns with its intrinsic knowledge. Additionally, TrustScore can seam"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.12545","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/2402.12545/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:16:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"15KqDtNlnNDHeDjnOK2x5Nrxc/xsSfzVnMSr1ixJhUzxb3NyaloLBpCzdqkOBC63oaJL1YMm64Ue7I/n3w3JAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T21:01:50.221842Z"},"content_sha256":"3607aa14996b60a5bdc77e427eac4159a0c8de9d914f26716544b26c016cc5cc","schema_version":"1.0","event_id":"sha256:3607aa14996b60a5bdc77e427eac4159a0c8de9d914f26716544b26c016cc5cc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NWPPO5TYZFU4WK2LCOM2UJYUBI/bundle.json","state_url":"https://pith.science/pith/NWPPO5TYZFU4WK2LCOM2UJYUBI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NWPPO5TYZFU4WK2LCOM2UJYUBI/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-06T21:01:50Z","links":{"resolver":"https://pith.science/pith/NWPPO5TYZFU4WK2LCOM2UJYUBI","bundle":"https://pith.science/pith/NWPPO5TYZFU4WK2LCOM2UJYUBI/bundle.json","state":"https://pith.science/pith/NWPPO5TYZFU4WK2LCOM2UJYUBI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NWPPO5TYZFU4WK2LCOM2UJYUBI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:NWPPO5TYZFU4WK2LCOM2UJYUBI","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"1726be6dc8e15628de56f4cbe6a98462a7d4c4728f90dc9ed1cf99a241f987da","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-19T21:12:14Z","title_canon_sha256":"b52d139339ca25081f1d97645f67949d4945ce3138bde84294598d0fd9237745"},"schema_version":"1.0","source":{"id":"2402.12545","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.12545","created_at":"2026-07-05T08:16:08Z"},{"alias_kind":"arxiv_version","alias_value":"2402.12545v2","created_at":"2026-07-05T08:16:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.12545","created_at":"2026-07-05T08:16:08Z"},{"alias_kind":"pith_short_12","alias_value":"NWPPO5TYZFU4","created_at":"2026-07-05T08:16:08Z"},{"alias_kind":"pith_short_16","alias_value":"NWPPO5TYZFU4WK2L","created_at":"2026-07-05T08:16:08Z"},{"alias_kind":"pith_short_8","alias_value":"NWPPO5TY","created_at":"2026-07-05T08:16:08Z"}],"graph_snapshots":[{"event_id":"sha256:3607aa14996b60a5bdc77e427eac4159a0c8de9d914f26716544b26c016cc5cc","target":"graph","created_at":"2026-07-05T08:16:08Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2402.12545/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) have demonstrated impressive capabilities across various domains, prompting a surge in their practical applications. However, concerns have arisen regarding the trustworthiness of LLMs outputs, particularly in closed-book question-answering tasks, where non-experts may struggle to identify inaccuracies due to the absence of contextual or ground truth information. This paper introduces TrustScore, a framework based on the concept of Behavioral Consistency, which evaluates whether an LLMs response aligns with its intrinsic knowledge. Additionally, TrustScore can seam","authors_text":"Danna Zheng, Danyang Liu, Jeff Z. Pan, Mirella Lapata","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-19T21:12:14Z","title":"TrustScore: Reference-Free Evaluation of LLM Response Trustworthiness"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.12545","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:a7988d792c9d36af1e91e9c0a12e089b0f2b715e9c590926172a00131ce9ff43","target":"record","created_at":"2026-07-05T08:16:08Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"1726be6dc8e15628de56f4cbe6a98462a7d4c4728f90dc9ed1cf99a241f987da","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-19T21:12:14Z","title_canon_sha256":"b52d139339ca25081f1d97645f67949d4945ce3138bde84294598d0fd9237745"},"schema_version":"1.0","source":{"id":"2402.12545","kind":"arxiv","version":2}},"canonical_sha256":"6d9ef77678c969cb2b4b1399aa27140a293ecfdb764e85dc409ba30541809d64","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6d9ef77678c969cb2b4b1399aa27140a293ecfdb764e85dc409ba30541809d64","first_computed_at":"2026-07-05T08:16:08.637218Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:16:08.637218Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fO8nbWnd8lm5L7I+ODs5FSbthhM52dDdDXKHPYjWpqFmuNtkTJBk8/soP7XbKPFIBLzLQzvbwe1Bk/YgIDTQAw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:16:08.637692Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.12545","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a7988d792c9d36af1e91e9c0a12e089b0f2b715e9c590926172a00131ce9ff43","sha256:3607aa14996b60a5bdc77e427eac4159a0c8de9d914f26716544b26c016cc5cc"],"state_sha256":"1df16b4704ba275b31c5f059cee131487e26f9dcef02be905ab64123c2f4449d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KiujOoT4aUkh3dZq2GNEVpXtr7dOKrUf4ykTgQImRIHZgEHtawreB5w0tD5OTzt/CG9uyKS55gFQu0zSxnrACA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T21:01:50.226734Z","bundle_sha256":"002d600b11b818c5a1d6bf8f935893333fd4fa7777ef5fd39e97a5dc32d20f05"}}