{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:LFZCVOGEPU2F3CX4ANVXGR5NRM","short_pith_number":"pith:LFZCVOGE","canonical_record":{"source":{"id":"2506.01114","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-01T18:42:24Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"602ac1b73a0b8de2cc4e03609f4ac058f2d802b5f384ef0a5cdf90b4e5389658","abstract_canon_sha256":"97d771e4a986e918b5910ed4789ceb616bf6057da42f5f6bb5d3d2efacfc2a34"},"schema_version":"1.0"},"canonical_sha256":"59722ab8c47d345d8afc036b7347ad8b379dd83b34e7894a32b3cd90bd54231d","source":{"kind":"arxiv","id":"2506.01114","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.01114","created_at":"2026-07-05T11:13:46Z"},{"alias_kind":"arxiv_version","alias_value":"2506.01114v1","created_at":"2026-07-05T11:13:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.01114","created_at":"2026-07-05T11:13:46Z"},{"alias_kind":"pith_short_12","alias_value":"LFZCVOGEPU2F","created_at":"2026-07-05T11:13:46Z"},{"alias_kind":"pith_short_16","alias_value":"LFZCVOGEPU2F3CX4","created_at":"2026-07-05T11:13:46Z"},{"alias_kind":"pith_short_8","alias_value":"LFZCVOGE","created_at":"2026-07-05T11:13:46Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:LFZCVOGEPU2F3CX4ANVXGR5NRM","target":"record","payload":{"canonical_record":{"source":{"id":"2506.01114","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-01T18:42:24Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"602ac1b73a0b8de2cc4e03609f4ac058f2d802b5f384ef0a5cdf90b4e5389658","abstract_canon_sha256":"97d771e4a986e918b5910ed4789ceb616bf6057da42f5f6bb5d3d2efacfc2a34"},"schema_version":"1.0"},"canonical_sha256":"59722ab8c47d345d8afc036b7347ad8b379dd83b34e7894a32b3cd90bd54231d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:13:46.667898Z","signature_b64":"bsw6UhFp6WNFjPhQcmirgyPt/U5HXa9QzWaS0NjIKdIB3ILh+i++YL9ad2E6UlH94j3dX07w5E8eCN2xTuwwBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"59722ab8c47d345d8afc036b7347ad8b379dd83b34e7894a32b3cd90bd54231d","last_reissued_at":"2026-07-05T11:13:46.667450Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:13:46.667450Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.01114","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:13:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UJv8VnXxXlMwGnyQMvP9xYZUMY2gCCT334Gi2P4MxuabP+Wul9g0qm//SNIVMPL2qHCPbGQgivU9q0HO+QiNCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T09:39:22.336837Z"},"content_sha256":"1cf783eb6218e7c321d2e9ac3e9439005e2e5b34c15b87343dd99cf123db3ea8","schema_version":"1.0","event_id":"sha256:1cf783eb6218e7c321d2e9ac3e9439005e2e5b34c15b87343dd99cf123db3ea8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:LFZCVOGEPU2F3CX4ANVXGR5NRM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Reconsidering LLM Uncertainty Estimation Methods in the Wild","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Baturalp Buyukates, Duygu Nur Yaldiz, Sai Praneeth Karimireddy, Salman Avestimehr, Sungmin Kang, Tuo Zhang, Yavuz Bakman","submitted_at":"2025-06-01T18:42:24Z","abstract_excerpt":"Large Language Model (LLM) Uncertainty Estimation (UE) methods have become a crucial tool for detecting hallucinations in recent years. While numerous UE methods have been proposed, most existing studies evaluate them in isolated short-form QA settings using threshold-independent metrics such as AUROC or PRR. However, real-world deployment of UE methods introduces several challenges. In this work, we systematically examine four key aspects of deploying UE methods in practical settings. Specifically, we assess (1) the sensitivity of UE methods to decision threshold selection, (2) their robustne"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.01114","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/2506.01114/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":"2026-07-05T11:13:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7qgf0O9SAXnumEXAgMGiw69U6bSPDMeiCVs8QcS+JOFZMWYWXW0f1W4x/eXGmjwulsN4yMPuS0Dc9a8G69DOAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T09:39:22.337422Z"},"content_sha256":"72d6b79fe9527dd368905b0c27456ffaf82d6ae91fe24922949c7de0f6f6cb53","schema_version":"1.0","event_id":"sha256:72d6b79fe9527dd368905b0c27456ffaf82d6ae91fe24922949c7de0f6f6cb53"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LFZCVOGEPU2F3CX4ANVXGR5NRM/bundle.json","state_url":"https://pith.science/pith/LFZCVOGEPU2F3CX4ANVXGR5NRM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LFZCVOGEPU2F3CX4ANVXGR5NRM/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":"2026-08-08T09:39:22Z","links":{"resolver":"https://pith.science/pith/LFZCVOGEPU2F3CX4ANVXGR5NRM","bundle":"https://pith.science/pith/LFZCVOGEPU2F3CX4ANVXGR5NRM/bundle.json","state":"https://pith.science/pith/LFZCVOGEPU2F3CX4ANVXGR5NRM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LFZCVOGEPU2F3CX4ANVXGR5NRM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:LFZCVOGEPU2F3CX4ANVXGR5NRM","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":"97d771e4a986e918b5910ed4789ceb616bf6057da42f5f6bb5d3d2efacfc2a34","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-01T18:42:24Z","title_canon_sha256":"602ac1b73a0b8de2cc4e03609f4ac058f2d802b5f384ef0a5cdf90b4e5389658"},"schema_version":"1.0","source":{"id":"2506.01114","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.01114","created_at":"2026-07-05T11:13:46Z"},{"alias_kind":"arxiv_version","alias_value":"2506.01114v1","created_at":"2026-07-05T11:13:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.01114","created_at":"2026-07-05T11:13:46Z"},{"alias_kind":"pith_short_12","alias_value":"LFZCVOGEPU2F","created_at":"2026-07-05T11:13:46Z"},{"alias_kind":"pith_short_16","alias_value":"LFZCVOGEPU2F3CX4","created_at":"2026-07-05T11:13:46Z"},{"alias_kind":"pith_short_8","alias_value":"LFZCVOGE","created_at":"2026-07-05T11:13:46Z"}],"graph_snapshots":[{"event_id":"sha256:72d6b79fe9527dd368905b0c27456ffaf82d6ae91fe24922949c7de0f6f6cb53","target":"graph","created_at":"2026-07-05T11:13:46Z","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/2506.01114/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Model (LLM) Uncertainty Estimation (UE) methods have become a crucial tool for detecting hallucinations in recent years. While numerous UE methods have been proposed, most existing studies evaluate them in isolated short-form QA settings using threshold-independent metrics such as AUROC or PRR. However, real-world deployment of UE methods introduces several challenges. In this work, we systematically examine four key aspects of deploying UE methods in practical settings. Specifically, we assess (1) the sensitivity of UE methods to decision threshold selection, (2) their robustne","authors_text":"Baturalp Buyukates, Duygu Nur Yaldiz, Sai Praneeth Karimireddy, Salman Avestimehr, Sungmin Kang, Tuo Zhang, Yavuz Bakman","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-01T18:42:24Z","title":"Reconsidering LLM Uncertainty Estimation Methods in the Wild"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.01114","kind":"arxiv","version":1},"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":"sha256:1cf783eb6218e7c321d2e9ac3e9439005e2e5b34c15b87343dd99cf123db3ea8","target":"record","created_at":"2026-07-05T11:13:46Z","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":"97d771e4a986e918b5910ed4789ceb616bf6057da42f5f6bb5d3d2efacfc2a34","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-01T18:42:24Z","title_canon_sha256":"602ac1b73a0b8de2cc4e03609f4ac058f2d802b5f384ef0a5cdf90b4e5389658"},"schema_version":"1.0","source":{"id":"2506.01114","kind":"arxiv","version":1}},"canonical_sha256":"59722ab8c47d345d8afc036b7347ad8b379dd83b34e7894a32b3cd90bd54231d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"59722ab8c47d345d8afc036b7347ad8b379dd83b34e7894a32b3cd90bd54231d","first_computed_at":"2026-07-05T11:13:46.667450Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:13:46.667450Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"bsw6UhFp6WNFjPhQcmirgyPt/U5HXa9QzWaS0NjIKdIB3ILh+i++YL9ad2E6UlH94j3dX07w5E8eCN2xTuwwBA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:13:46.667898Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.01114","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1cf783eb6218e7c321d2e9ac3e9439005e2e5b34c15b87343dd99cf123db3ea8","sha256:72d6b79fe9527dd368905b0c27456ffaf82d6ae91fe24922949c7de0f6f6cb53"],"state_sha256":"43bc1a882285231037f2da6e791dd43908eb3097df90fe0de9c368e40926d28f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"urCziI3AoP/nZEI0NNd+wB8HPlU6Y3J3VCqhkyLN9p7rwyqadwLoxW6I+eMhADWFla0b0BJL8vL8SolprC4lDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T09:39:22.341194Z","bundle_sha256":"37e1518ef3fc140e087a43060d503d4fd4853d385df91a0efcb1e94dd79c828d"}}