{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:6HGETMJUF7Q7VHZ5V73E55MVW7","short_pith_number":"pith:6HGETMJU","schema_version":"1.0","canonical_sha256":"f1cc49b1342fe1fa9f3daff64ef595b7d70021e7b60e6675ec6bf1d670e0744d","source":{"kind":"arxiv","id":"2506.02696","version":1},"attestation_state":"computed","paper":{"title":"Shaking to Reveal: Perturbation-Based Detection of LLM Hallucinations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Jinyuan Luo, Ling Chen, Seongheon Park, Yixuan Li, Zhen Fang","submitted_at":"2025-06-03T09:44:28Z","abstract_excerpt":"Hallucination remains a key obstacle to the reliable deployment of large language models (LLMs) in real-world question answering tasks. A widely adopted strategy to detect hallucination, known as self-assessment, relies on the model's own output confidence to estimate the factual accuracy of its answers. However, this strategy assumes that the model's output distribution closely reflects the true data distribution, which may not always hold in practice. As bias accumulates through the model's layers, the final output can diverge from the underlying reasoning process, making output-level confid"},"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":"2506.02696","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-06-03T09:44:28Z","cross_cats_sorted":[],"title_canon_sha256":"66a01147f9771ffa142aa4f1ee1e7be4777367fbd962cfcfd928c6f29c2100e7","abstract_canon_sha256":"0ed0eac1cc7dc7d1fddcc8b2bee763b8636e03d9e574fafc86c9e83cb2106644"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:15:06.724685Z","signature_b64":"Y7PUGW5SZ+EKgSTdkQcTTjF6D9dfLFttk9zOxzbT8gO6ggNzC5D+aVFQIaKA6CqxJBcm8NfPxAlGbgCqoq7TDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f1cc49b1342fe1fa9f3daff64ef595b7d70021e7b60e6675ec6bf1d670e0744d","last_reissued_at":"2026-07-05T11:15:06.724165Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:15:06.724165Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Shaking to Reveal: Perturbation-Based Detection of LLM Hallucinations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Jinyuan Luo, Ling Chen, Seongheon Park, Yixuan Li, Zhen Fang","submitted_at":"2025-06-03T09:44:28Z","abstract_excerpt":"Hallucination remains a key obstacle to the reliable deployment of large language models (LLMs) in real-world question answering tasks. A widely adopted strategy to detect hallucination, known as self-assessment, relies on the model's own output confidence to estimate the factual accuracy of its answers. However, this strategy assumes that the model's output distribution closely reflects the true data distribution, which may not always hold in practice. As bias accumulates through the model's layers, the final output can diverge from the underlying reasoning process, making output-level confid"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.02696","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/2506.02696/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":"2506.02696","created_at":"2026-07-05T11:15:06.724234+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.02696v1","created_at":"2026-07-05T11:15:06.724234+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.02696","created_at":"2026-07-05T11:15:06.724234+00:00"},{"alias_kind":"pith_short_12","alias_value":"6HGETMJUF7Q7","created_at":"2026-07-05T11:15:06.724234+00:00"},{"alias_kind":"pith_short_16","alias_value":"6HGETMJUF7Q7VHZ5","created_at":"2026-07-05T11:15:06.724234+00:00"},{"alias_kind":"pith_short_8","alias_value":"6HGETMJU","created_at":"2026-07-05T11:15:06.724234+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/6HGETMJUF7Q7VHZ5V73E55MVW7","json":"https://pith.science/pith/6HGETMJUF7Q7VHZ5V73E55MVW7.json","graph_json":"https://pith.science/api/pith-number/6HGETMJUF7Q7VHZ5V73E55MVW7/graph.json","events_json":"https://pith.science/api/pith-number/6HGETMJUF7Q7VHZ5V73E55MVW7/events.json","paper":"https://pith.science/paper/6HGETMJU"},"agent_actions":{"view_html":"https://pith.science/pith/6HGETMJUF7Q7VHZ5V73E55MVW7","download_json":"https://pith.science/pith/6HGETMJUF7Q7VHZ5V73E55MVW7.json","view_paper":"https://pith.science/paper/6HGETMJU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.02696&json=true","fetch_graph":"https://pith.science/api/pith-number/6HGETMJUF7Q7VHZ5V73E55MVW7/graph.json","fetch_events":"https://pith.science/api/pith-number/6HGETMJUF7Q7VHZ5V73E55MVW7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6HGETMJUF7Q7VHZ5V73E55MVW7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6HGETMJUF7Q7VHZ5V73E55MVW7/action/storage_attestation","attest_author":"https://pith.science/pith/6HGETMJUF7Q7VHZ5V73E55MVW7/action/author_attestation","sign_citation":"https://pith.science/pith/6HGETMJUF7Q7VHZ5V73E55MVW7/action/citation_signature","submit_replication":"https://pith.science/pith/6HGETMJUF7Q7VHZ5V73E55MVW7/action/replication_record"}},"created_at":"2026-07-05T11:15:06.724234+00:00","updated_at":"2026-07-05T11:15:06.724234+00:00"}