{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:7PZYTFEGK757CBTUYVZBNQHJ3K","short_pith_number":"pith:7PZYTFEG","canonical_record":{"source":{"id":"2601.08064","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-01-12T23:16:50Z","cross_cats_sorted":[],"title_canon_sha256":"660cb0435497d4ff6e65e8b8b1bc869edb4c3f96227410ff1f8aa7bad4b1837e","abstract_canon_sha256":"e3206c149382da2e970d36fb09a66a54c4e48f4576ebca05f0a75c81f4606fa0"},"schema_version":"1.0"},"canonical_sha256":"fbf389948657fbf10674c57216c0e9da885d0f9b686e4efb86d8fbc7f7718747","source":{"kind":"arxiv","id":"2601.08064","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2601.08064","created_at":"2026-05-29T01:05:01Z"},{"alias_kind":"arxiv_version","alias_value":"2601.08064v2","created_at":"2026-05-29T01:05:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.08064","created_at":"2026-05-29T01:05:01Z"},{"alias_kind":"pith_short_12","alias_value":"7PZYTFEGK757","created_at":"2026-05-29T01:05:01Z"},{"alias_kind":"pith_short_16","alias_value":"7PZYTFEGK757CBTU","created_at":"2026-05-29T01:05:01Z"},{"alias_kind":"pith_short_8","alias_value":"7PZYTFEG","created_at":"2026-05-29T01:05:01Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:7PZYTFEGK757CBTUYVZBNQHJ3K","target":"record","payload":{"canonical_record":{"source":{"id":"2601.08064","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-01-12T23:16:50Z","cross_cats_sorted":[],"title_canon_sha256":"660cb0435497d4ff6e65e8b8b1bc869edb4c3f96227410ff1f8aa7bad4b1837e","abstract_canon_sha256":"e3206c149382da2e970d36fb09a66a54c4e48f4576ebca05f0a75c81f4606fa0"},"schema_version":"1.0"},"canonical_sha256":"fbf389948657fbf10674c57216c0e9da885d0f9b686e4efb86d8fbc7f7718747","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:05:01.773973Z","signature_b64":"k6RQPpGQgK5QWFRJvFr6WxCtsuO3pgjdj254sOc5ZHQVhuh1cq1HfkDavwqTVoRNP+OQ4oUa0PX7tlC0bpx6Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fbf389948657fbf10674c57216c0e9da885d0f9b686e4efb86d8fbc7f7718747","last_reissued_at":"2026-05-29T01:05:01.773093Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:05:01.773093Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2601.08064","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":"2026-05-29T01:05:01Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"b/1gKYMoMF8flskXK/g9czpTeM5Bp778YlftdXuw+EgQRz7vOObWVQZyggwa5LYlygqrHQrXKPg6DsMn5VPYAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-02T18:43:50.684055Z"},"content_sha256":"d8995ca5a28cca3cc982eabc6a84307030d60a237d42f97fd991793e4a3282f9","schema_version":"1.0","event_id":"sha256:d8995ca5a28cca3cc982eabc6a84307030d60a237d42f97fd991793e4a3282f9"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:7PZYTFEGK757CBTUYVZBNQHJ3K","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Calibration Is Not Enough: Evaluating Confidence Estimation Under Language Variations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Benjamin Roth, Dennis Ulmer, Hinrich Sch\\\"utze, Terra Blevins, Yihong Liu, Yuxi Xia","submitted_at":"2026-01-12T23:16:50Z","abstract_excerpt":"Confidence estimation (CE) indicates how reliable the answers of large language models are and impacts user trust and decision-making. Existing evaluations mainly concern the alignment between confidence and correctness, but ignore the variability of language: confidence estimates should remain consistent under semantically equivalent prompts or answer variations, while changing when answer meaning differs, as this may indicate a change in correctness. Therefore, we introduce a novel evaluation framework based on three complementary properties: \\textbf{robustness} to prompt perturbations, \\tex"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.08064","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/2601.08064/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":"2026-05-29T01:05:01Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8cthIK9biohc8BIPJSV9TWgIQ3pfVN9U/QXWhCu8azein6p+EhUCRUUUE9OCtjonmmIOgMd2OIjplU96+s+bCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-02T18:43:50.684462Z"},"content_sha256":"962e63d8b877653c205e45bc935d691490efc9d82df8655268841cf23a61c48a","schema_version":"1.0","event_id":"sha256:962e63d8b877653c205e45bc935d691490efc9d82df8655268841cf23a61c48a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7PZYTFEGK757CBTUYVZBNQHJ3K/bundle.json","state_url":"https://pith.science/pith/7PZYTFEGK757CBTUYVZBNQHJ3K/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7PZYTFEGK757CBTUYVZBNQHJ3K/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":"2026-06-02T18:43:50Z","links":{"resolver":"https://pith.science/pith/7PZYTFEGK757CBTUYVZBNQHJ3K","bundle":"https://pith.science/pith/7PZYTFEGK757CBTUYVZBNQHJ3K/bundle.json","state":"https://pith.science/pith/7PZYTFEGK757CBTUYVZBNQHJ3K/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7PZYTFEGK757CBTUYVZBNQHJ3K/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:7PZYTFEGK757CBTUYVZBNQHJ3K","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":"e3206c149382da2e970d36fb09a66a54c4e48f4576ebca05f0a75c81f4606fa0","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-01-12T23:16:50Z","title_canon_sha256":"660cb0435497d4ff6e65e8b8b1bc869edb4c3f96227410ff1f8aa7bad4b1837e"},"schema_version":"1.0","source":{"id":"2601.08064","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2601.08064","created_at":"2026-05-29T01:05:01Z"},{"alias_kind":"arxiv_version","alias_value":"2601.08064v2","created_at":"2026-05-29T01:05:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.08064","created_at":"2026-05-29T01:05:01Z"},{"alias_kind":"pith_short_12","alias_value":"7PZYTFEGK757","created_at":"2026-05-29T01:05:01Z"},{"alias_kind":"pith_short_16","alias_value":"7PZYTFEGK757CBTU","created_at":"2026-05-29T01:05:01Z"},{"alias_kind":"pith_short_8","alias_value":"7PZYTFEG","created_at":"2026-05-29T01:05:01Z"}],"graph_snapshots":[{"event_id":"sha256:962e63d8b877653c205e45bc935d691490efc9d82df8655268841cf23a61c48a","target":"graph","created_at":"2026-05-29T01:05:01Z","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/2601.08064/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Confidence estimation (CE) indicates how reliable the answers of large language models are and impacts user trust and decision-making. Existing evaluations mainly concern the alignment between confidence and correctness, but ignore the variability of language: confidence estimates should remain consistent under semantically equivalent prompts or answer variations, while changing when answer meaning differs, as this may indicate a change in correctness. Therefore, we introduce a novel evaluation framework based on three complementary properties: \\textbf{robustness} to prompt perturbations, \\tex","authors_text":"Benjamin Roth, Dennis Ulmer, Hinrich Sch\\\"utze, Terra Blevins, Yihong Liu, Yuxi Xia","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-01-12T23:16:50Z","title":"Calibration Is Not Enough: Evaluating Confidence Estimation Under Language Variations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.08064","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:d8995ca5a28cca3cc982eabc6a84307030d60a237d42f97fd991793e4a3282f9","target":"record","created_at":"2026-05-29T01:05:01Z","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":"e3206c149382da2e970d36fb09a66a54c4e48f4576ebca05f0a75c81f4606fa0","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-01-12T23:16:50Z","title_canon_sha256":"660cb0435497d4ff6e65e8b8b1bc869edb4c3f96227410ff1f8aa7bad4b1837e"},"schema_version":"1.0","source":{"id":"2601.08064","kind":"arxiv","version":2}},"canonical_sha256":"fbf389948657fbf10674c57216c0e9da885d0f9b686e4efb86d8fbc7f7718747","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fbf389948657fbf10674c57216c0e9da885d0f9b686e4efb86d8fbc7f7718747","first_computed_at":"2026-05-29T01:05:01.773093Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-29T01:05:01.773093Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"k6RQPpGQgK5QWFRJvFr6WxCtsuO3pgjdj254sOc5ZHQVhuh1cq1HfkDavwqTVoRNP+OQ4oUa0PX7tlC0bpx6Aw==","signature_status":"signed_v1","signed_at":"2026-05-29T01:05:01.773973Z","signed_message":"canonical_sha256_bytes"},"source_id":"2601.08064","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d8995ca5a28cca3cc982eabc6a84307030d60a237d42f97fd991793e4a3282f9","sha256:962e63d8b877653c205e45bc935d691490efc9d82df8655268841cf23a61c48a"],"state_sha256":"8e8ebe21f962d455abb9714d67c808282cf8589a94013aa5f4b001b7b46df881"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tIXFJcfO1VKMyu6aoAfUIcMOmvZkjGT3V98wXsZCShs9JojTVKQdOdKsiMzVkEXjllIpEI6FAlvp3yLRxzCBBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-02T18:43:50.686431Z","bundle_sha256":"0870f99f9dc1668fdde3cefac32d35045532b230ee2f52f8d0f83a310e13fe27"}}