{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:T33SUEZDO6CTC7YX4G3KUEIEYX","short_pith_number":"pith:T33SUEZD","schema_version":"1.0","canonical_sha256":"9ef72a13237785317f17e1b6aa1104c5ec9fb74592a1f0b8cd5a84ce95f2f14f","source":{"kind":"arxiv","id":"2608.10315","version":1},"attestation_state":"computed","paper":{"title":"Is This Your Final Answer? Cross-Contextual Consistency as a Measure of LLM Credibility","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bryon Aragam, Siyang Wu, Yibo Jiang","submitted_at":"2026-08-10T23:38:10Z","abstract_excerpt":"Large language models (LLMs) are powerful black-box systems, making it difficult to discern whether their answers reflect stable internal beliefs or superficial pattern matching. We identify cross-contextual consistency as an underutilized behavioral property of LLMs: a credible answer should remain stable when the same task is placed under topic-aligned, content-neutral contextual variation. Building on this intuition, we operationalize Cross-Contextual Consistency (C3) by comparing model generations under original and perturbed prompts. Across 26 models and six benchmarks spanning reasoning,"},"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":"2608.10315","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-10T23:38:10Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d0f03172d0aba26e89f35e6fa92bfb25cc26fe5f132a74c47bd0a1ee0a7a7da9","abstract_canon_sha256":"3186dc94e9d15aff8c9d927f5123b549fa8adf8164153ebb426d13a628a8cf57"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-12T00:23:51.716990Z","signature_b64":"j/XGF/cHmSBDjks+d72aI3H6rki8LWz6g3iWBb8xlO9UGeNyxF/XdYFHxYw9CBlZ8jtVZnu5koclch3uVPLFCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9ef72a13237785317f17e1b6aa1104c5ec9fb74592a1f0b8cd5a84ce95f2f14f","last_reissued_at":"2026-08-12T00:23:51.715432Z","signature_status":"signed_v1","first_computed_at":"2026-08-12T00:23:51.715432Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Is This Your Final Answer? Cross-Contextual Consistency as a Measure of LLM Credibility","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bryon Aragam, Siyang Wu, Yibo Jiang","submitted_at":"2026-08-10T23:38:10Z","abstract_excerpt":"Large language models (LLMs) are powerful black-box systems, making it difficult to discern whether their answers reflect stable internal beliefs or superficial pattern matching. We identify cross-contextual consistency as an underutilized behavioral property of LLMs: a credible answer should remain stable when the same task is placed under topic-aligned, content-neutral contextual variation. Building on this intuition, we operationalize Cross-Contextual Consistency (C3) by comparing model generations under original and perturbed prompts. Across 26 models and six benchmarks spanning reasoning,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.10315","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/2608.10315/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":"2608.10315","created_at":"2026-08-12T00:23:51.718172+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.10315v1","created_at":"2026-08-12T00:23:51.718172+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.10315","created_at":"2026-08-12T00:23:51.718172+00:00"},{"alias_kind":"pith_short_12","alias_value":"T33SUEZDO6CT","created_at":"2026-08-12T00:23:51.718172+00:00"},{"alias_kind":"pith_short_16","alias_value":"T33SUEZDO6CTC7YX","created_at":"2026-08-12T00:23:51.718172+00:00"},{"alias_kind":"pith_short_8","alias_value":"T33SUEZD","created_at":"2026-08-12T00:23:51.718172+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/T33SUEZDO6CTC7YX4G3KUEIEYX","json":"https://pith.science/pith/T33SUEZDO6CTC7YX4G3KUEIEYX.json","graph_json":"https://pith.science/api/pith-number/T33SUEZDO6CTC7YX4G3KUEIEYX/graph.json","events_json":"https://pith.science/api/pith-number/T33SUEZDO6CTC7YX4G3KUEIEYX/events.json","paper":"https://pith.science/paper/T33SUEZD"},"agent_actions":{"view_html":"https://pith.science/pith/T33SUEZDO6CTC7YX4G3KUEIEYX","download_json":"https://pith.science/pith/T33SUEZDO6CTC7YX4G3KUEIEYX.json","view_paper":"https://pith.science/paper/T33SUEZD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.10315&json=true","fetch_graph":"https://pith.science/api/pith-number/T33SUEZDO6CTC7YX4G3KUEIEYX/graph.json","fetch_events":"https://pith.science/api/pith-number/T33SUEZDO6CTC7YX4G3KUEIEYX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/T33SUEZDO6CTC7YX4G3KUEIEYX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/T33SUEZDO6CTC7YX4G3KUEIEYX/action/storage_attestation","attest_author":"https://pith.science/pith/T33SUEZDO6CTC7YX4G3KUEIEYX/action/author_attestation","sign_citation":"https://pith.science/pith/T33SUEZDO6CTC7YX4G3KUEIEYX/action/citation_signature","submit_replication":"https://pith.science/pith/T33SUEZDO6CTC7YX4G3KUEIEYX/action/replication_record"}},"created_at":"2026-08-12T00:23:51.718172+00:00","updated_at":"2026-08-12T00:23:51.718172+00:00"}