{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:IGBE4MDUSU4SG5WYNMF33H5CYV","short_pith_number":"pith:IGBE4MDU","canonical_record":{"source":{"id":"2505.21701","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-27T19:39:49Z","cross_cats_sorted":[],"title_canon_sha256":"60482ded6518c5c747c5f5c249f5bfba4c434c900da125046cc80eef9cb1a20b","abstract_canon_sha256":"7b26d6027750db789f95d068e6d7e9a0d930e24f7e87ac4f08e5497c325284e9"},"schema_version":"1.0"},"canonical_sha256":"41824e307495392376d86b0bbd9fa2c552f1bc9f2672787d24883628ad731781","source":{"kind":"arxiv","id":"2505.21701","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.21701","created_at":"2026-07-05T11:12:33Z"},{"alias_kind":"arxiv_version","alias_value":"2505.21701v2","created_at":"2026-07-05T11:12:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.21701","created_at":"2026-07-05T11:12:33Z"},{"alias_kind":"pith_short_12","alias_value":"IGBE4MDUSU4S","created_at":"2026-07-05T11:12:33Z"},{"alias_kind":"pith_short_16","alias_value":"IGBE4MDUSU4SG5WY","created_at":"2026-07-05T11:12:33Z"},{"alias_kind":"pith_short_8","alias_value":"IGBE4MDU","created_at":"2026-07-05T11:12:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:IGBE4MDUSU4SG5WYNMF33H5CYV","target":"record","payload":{"canonical_record":{"source":{"id":"2505.21701","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-27T19:39:49Z","cross_cats_sorted":[],"title_canon_sha256":"60482ded6518c5c747c5f5c249f5bfba4c434c900da125046cc80eef9cb1a20b","abstract_canon_sha256":"7b26d6027750db789f95d068e6d7e9a0d930e24f7e87ac4f08e5497c325284e9"},"schema_version":"1.0"},"canonical_sha256":"41824e307495392376d86b0bbd9fa2c552f1bc9f2672787d24883628ad731781","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:33.816963Z","signature_b64":"7cYOtHtTAUl+Pk1/Gyn0GfyzhtXy1MoUMEVT9cZfsdQEhQyQSIMMU9wKnQJBqveBsCE1l9YDZA6Mzw+mDGSAAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"41824e307495392376d86b0bbd9fa2c552f1bc9f2672787d24883628ad731781","last_reissued_at":"2026-07-05T11:12:33.816358Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:33.816358Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.21701","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":"2026-07-05T11:12:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mkaxYcvIIR+WYK5X3vsXnTC752rT97Vce8mJyVQwRx7Br4atFVkwoJ3pL3ZGHJglry+uqhmk/QGOIrkPX6faAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T03:13:28.148968Z"},"content_sha256":"6554a9888707d5a949e941c7a008f0edc3bd75059bd611d28a3d84633c553893","schema_version":"1.0","event_id":"sha256:6554a9888707d5a949e941c7a008f0edc3bd75059bd611d28a3d84633c553893"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:IGBE4MDUSU4SG5WYNMF33H5CYV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Do We Know What LLMs Don't Know? A Study of Consistency in Knowledge Probing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Abdullatif K\\\"oksal, Ali Modarressi, Hinrich Sch\\\"utze, Michael A. Hedderich, Raoyuan Zhao","submitted_at":"2025-05-27T19:39:49Z","abstract_excerpt":"The reliability of large language models (LLMs) is greatly compromised by their tendency to hallucinate, underscoring the need for precise identification of knowledge gaps within LLMs. Various methods for probing such gaps exist, ranging from calibration-based to prompting-based methods. To evaluate these probing methods, in this paper, we propose a new process based on using input variations and quantitative metrics. Through this, we expose two dimensions of inconsistency in knowledge gap probing. (1) Intra-method inconsistency: Minimal non-semantic perturbations in prompts lead to considerab"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.21701","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/2505.21701/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":"2026-07-05T11:12:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BQltWf7lnsgdQv6Lp1x+Tn9HLJSSIh/CbtQnBN68oi83Bvd4QP/eaHqRU44Qos6uMDN3SPsiWCV3z+OyIh2FAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T03:13:28.149478Z"},"content_sha256":"5496a2a5cd6c8e4fdc8bea225aa0907f16b5a7d9569749f4c24d191ba3eabcf0","schema_version":"1.0","event_id":"sha256:5496a2a5cd6c8e4fdc8bea225aa0907f16b5a7d9569749f4c24d191ba3eabcf0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/IGBE4MDUSU4SG5WYNMF33H5CYV/bundle.json","state_url":"https://pith.science/pith/IGBE4MDUSU4SG5WYNMF33H5CYV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/IGBE4MDUSU4SG5WYNMF33H5CYV/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":"2026-08-10T03:13:28Z","links":{"resolver":"https://pith.science/pith/IGBE4MDUSU4SG5WYNMF33H5CYV","bundle":"https://pith.science/pith/IGBE4MDUSU4SG5WYNMF33H5CYV/bundle.json","state":"https://pith.science/pith/IGBE4MDUSU4SG5WYNMF33H5CYV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/IGBE4MDUSU4SG5WYNMF33H5CYV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:IGBE4MDUSU4SG5WYNMF33H5CYV","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":"7b26d6027750db789f95d068e6d7e9a0d930e24f7e87ac4f08e5497c325284e9","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-27T19:39:49Z","title_canon_sha256":"60482ded6518c5c747c5f5c249f5bfba4c434c900da125046cc80eef9cb1a20b"},"schema_version":"1.0","source":{"id":"2505.21701","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.21701","created_at":"2026-07-05T11:12:33Z"},{"alias_kind":"arxiv_version","alias_value":"2505.21701v2","created_at":"2026-07-05T11:12:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.21701","created_at":"2026-07-05T11:12:33Z"},{"alias_kind":"pith_short_12","alias_value":"IGBE4MDUSU4S","created_at":"2026-07-05T11:12:33Z"},{"alias_kind":"pith_short_16","alias_value":"IGBE4MDUSU4SG5WY","created_at":"2026-07-05T11:12:33Z"},{"alias_kind":"pith_short_8","alias_value":"IGBE4MDU","created_at":"2026-07-05T11:12:33Z"}],"graph_snapshots":[{"event_id":"sha256:5496a2a5cd6c8e4fdc8bea225aa0907f16b5a7d9569749f4c24d191ba3eabcf0","target":"graph","created_at":"2026-07-05T11:12:33Z","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/2505.21701/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The reliability of large language models (LLMs) is greatly compromised by their tendency to hallucinate, underscoring the need for precise identification of knowledge gaps within LLMs. Various methods for probing such gaps exist, ranging from calibration-based to prompting-based methods. To evaluate these probing methods, in this paper, we propose a new process based on using input variations and quantitative metrics. Through this, we expose two dimensions of inconsistency in knowledge gap probing. (1) Intra-method inconsistency: Minimal non-semantic perturbations in prompts lead to considerab","authors_text":"Abdullatif K\\\"oksal, Ali Modarressi, Hinrich Sch\\\"utze, Michael A. Hedderich, Raoyuan Zhao","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-27T19:39:49Z","title":"Do We Know What LLMs Don't Know? A Study of Consistency in Knowledge Probing"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.21701","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":"sha256:6554a9888707d5a949e941c7a008f0edc3bd75059bd611d28a3d84633c553893","target":"record","created_at":"2026-07-05T11:12:33Z","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":"7b26d6027750db789f95d068e6d7e9a0d930e24f7e87ac4f08e5497c325284e9","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-27T19:39:49Z","title_canon_sha256":"60482ded6518c5c747c5f5c249f5bfba4c434c900da125046cc80eef9cb1a20b"},"schema_version":"1.0","source":{"id":"2505.21701","kind":"arxiv","version":2}},"canonical_sha256":"41824e307495392376d86b0bbd9fa2c552f1bc9f2672787d24883628ad731781","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"41824e307495392376d86b0bbd9fa2c552f1bc9f2672787d24883628ad731781","first_computed_at":"2026-07-05T11:12:33.816358Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:12:33.816358Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7cYOtHtTAUl+Pk1/Gyn0GfyzhtXy1MoUMEVT9cZfsdQEhQyQSIMMU9wKnQJBqveBsCE1l9YDZA6Mzw+mDGSAAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:12:33.816963Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.21701","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6554a9888707d5a949e941c7a008f0edc3bd75059bd611d28a3d84633c553893","sha256:5496a2a5cd6c8e4fdc8bea225aa0907f16b5a7d9569749f4c24d191ba3eabcf0"],"state_sha256":"eb8df541257faa1f720d4e67661121ae3bc7dfb9de43b634ac910cabba7a9159"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"g2PgKDTHH75ZRyyMPZcTq0JxQjFgHNT6cYxPiyXkvbmbbOyMXn5aeQnBUQOPfYTMd6ZHG+uJE/+Xws2pG5ALCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T03:13:28.155463Z","bundle_sha256":"e17e300f404709d3443c3ca546fedf7125f705f4d44e6d5ea87479f7651fd365"}}