{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:CH7A2ZXYFSTTRJEVAR537VUMV6","short_pith_number":"pith:CH7A2ZXY","canonical_record":{"source":{"id":"2608.04899","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-05T14:24:34Z","cross_cats_sorted":[],"title_canon_sha256":"e32165c60c0fcc9f22ac1f67aa41bc5956e649875fcf951ad5c9a8585e34403d","abstract_canon_sha256":"ce88332c3aa473fbf557e690bd9b901abe49f9ede39e17ee48797d33ce2f558a"},"schema_version":"1.0"},"canonical_sha256":"11fe0d66f82ca738a495047bbfd68caf8cd04ed8074034ae50961be594122872","source":{"kind":"arxiv","id":"2608.04899","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.04899","created_at":"2026-08-06T01:47:39Z"},{"alias_kind":"arxiv_version","alias_value":"2608.04899v1","created_at":"2026-08-06T01:47:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.04899","created_at":"2026-08-06T01:47:39Z"},{"alias_kind":"pith_short_12","alias_value":"CH7A2ZXYFSTT","created_at":"2026-08-06T01:47:39Z"},{"alias_kind":"pith_short_16","alias_value":"CH7A2ZXYFSTTRJEV","created_at":"2026-08-06T01:47:39Z"},{"alias_kind":"pith_short_8","alias_value":"CH7A2ZXY","created_at":"2026-08-06T01:47:39Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:CH7A2ZXYFSTTRJEVAR537VUMV6","target":"record","payload":{"canonical_record":{"source":{"id":"2608.04899","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-05T14:24:34Z","cross_cats_sorted":[],"title_canon_sha256":"e32165c60c0fcc9f22ac1f67aa41bc5956e649875fcf951ad5c9a8585e34403d","abstract_canon_sha256":"ce88332c3aa473fbf557e690bd9b901abe49f9ede39e17ee48797d33ce2f558a"},"schema_version":"1.0"},"canonical_sha256":"11fe0d66f82ca738a495047bbfd68caf8cd04ed8074034ae50961be594122872","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-06T01:47:39.946643Z","signature_b64":"2xY7iu145e+QY0xaczgmSQu4hPlsgQDAJh0kKFmZ5z/uWozY/ct/LAPbhgeEJ8PazSA+hwfDBnOXcLt0vdd4Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"11fe0d66f82ca738a495047bbfd68caf8cd04ed8074034ae50961be594122872","last_reissued_at":"2026-08-06T01:47:39.945287Z","signature_status":"signed_v1","first_computed_at":"2026-08-06T01:47:39.945287Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2608.04899","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":"2026-08-06T01:47:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"h/8JN3Je7gBDyBThtBHHUIURfaTyJC/KrIEgXTKvyHRdDFTQPZG5lMz3R3tuOQWWNHujzbrzKvNxn/EPFhPhDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T12:26:44.318849Z"},"content_sha256":"4bfc248835499ee2e19eaae13bfeeaf6b4b1872f5dbfdfa553631238764c34eb","schema_version":"1.0","event_id":"sha256:4bfc248835499ee2e19eaae13bfeeaf6b4b1872f5dbfdfa553631238764c34eb"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:CH7A2ZXYFSTTRJEVAR537VUMV6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Evaluation Pitfalls and Sparsity Limitations in LLM-based Confidence Estimates for Classification","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Andreas R\\\"uckl\\'e, Elena Merdjanovska, Omar Zaidan","submitted_at":"2026-08-05T14:24:34Z","abstract_excerpt":"Confidence estimation is essential when LLMs are used for classification, indicating when predictions can be trusted. However, common approaches such as verbalization produce extremely sparse outputs. For instance, Qwen3-32B verbalizes only eight unique confidence values on SST-2, with over half being exactly 95%, a pattern we observe consistently across four datasets and two LLMs. Besides limiting practical utility, we show that this sparsity critically affects evaluation: the choice of interpolation in area under the accuracy-rejection curve (AUARC) dramatically alters rankings, with consist"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.04899","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/2608.04899/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":"2026-08-06T01:47:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yWoqOFnFZC+DOPe6y1NmMn6peuqO/t7ks2xdEOHFLiFFI/xC4qqjcuCDhWApR037f+tFamEyOPZqowijfwlACg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T12:26:44.319360Z"},"content_sha256":"826bd0e2ce9efc7695a7eeb2a43a41c50c97281128fa5f416ae30ad77715bd8f","schema_version":"1.0","event_id":"sha256:826bd0e2ce9efc7695a7eeb2a43a41c50c97281128fa5f416ae30ad77715bd8f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CH7A2ZXYFSTTRJEVAR537VUMV6/bundle.json","state_url":"https://pith.science/pith/CH7A2ZXYFSTTRJEVAR537VUMV6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CH7A2ZXYFSTTRJEVAR537VUMV6/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":"2026-08-09T12:26:44Z","links":{"resolver":"https://pith.science/pith/CH7A2ZXYFSTTRJEVAR537VUMV6","bundle":"https://pith.science/pith/CH7A2ZXYFSTTRJEVAR537VUMV6/bundle.json","state":"https://pith.science/pith/CH7A2ZXYFSTTRJEVAR537VUMV6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CH7A2ZXYFSTTRJEVAR537VUMV6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:CH7A2ZXYFSTTRJEVAR537VUMV6","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":"ce88332c3aa473fbf557e690bd9b901abe49f9ede39e17ee48797d33ce2f558a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-05T14:24:34Z","title_canon_sha256":"e32165c60c0fcc9f22ac1f67aa41bc5956e649875fcf951ad5c9a8585e34403d"},"schema_version":"1.0","source":{"id":"2608.04899","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.04899","created_at":"2026-08-06T01:47:39Z"},{"alias_kind":"arxiv_version","alias_value":"2608.04899v1","created_at":"2026-08-06T01:47:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.04899","created_at":"2026-08-06T01:47:39Z"},{"alias_kind":"pith_short_12","alias_value":"CH7A2ZXYFSTT","created_at":"2026-08-06T01:47:39Z"},{"alias_kind":"pith_short_16","alias_value":"CH7A2ZXYFSTTRJEV","created_at":"2026-08-06T01:47:39Z"},{"alias_kind":"pith_short_8","alias_value":"CH7A2ZXY","created_at":"2026-08-06T01:47:39Z"}],"graph_snapshots":[{"event_id":"sha256:826bd0e2ce9efc7695a7eeb2a43a41c50c97281128fa5f416ae30ad77715bd8f","target":"graph","created_at":"2026-08-06T01:47:39Z","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/2608.04899/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Confidence estimation is essential when LLMs are used for classification, indicating when predictions can be trusted. However, common approaches such as verbalization produce extremely sparse outputs. For instance, Qwen3-32B verbalizes only eight unique confidence values on SST-2, with over half being exactly 95%, a pattern we observe consistently across four datasets and two LLMs. Besides limiting practical utility, we show that this sparsity critically affects evaluation: the choice of interpolation in area under the accuracy-rejection curve (AUARC) dramatically alters rankings, with consist","authors_text":"Andreas R\\\"uckl\\'e, Elena Merdjanovska, Omar Zaidan","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-05T14:24:34Z","title":"Evaluation Pitfalls and Sparsity Limitations in LLM-based Confidence Estimates for Classification"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.04899","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":"sha256:4bfc248835499ee2e19eaae13bfeeaf6b4b1872f5dbfdfa553631238764c34eb","target":"record","created_at":"2026-08-06T01:47:39Z","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":"ce88332c3aa473fbf557e690bd9b901abe49f9ede39e17ee48797d33ce2f558a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-05T14:24:34Z","title_canon_sha256":"e32165c60c0fcc9f22ac1f67aa41bc5956e649875fcf951ad5c9a8585e34403d"},"schema_version":"1.0","source":{"id":"2608.04899","kind":"arxiv","version":1}},"canonical_sha256":"11fe0d66f82ca738a495047bbfd68caf8cd04ed8074034ae50961be594122872","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"11fe0d66f82ca738a495047bbfd68caf8cd04ed8074034ae50961be594122872","first_computed_at":"2026-08-06T01:47:39.945287Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-06T01:47:39.945287Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"2xY7iu145e+QY0xaczgmSQu4hPlsgQDAJh0kKFmZ5z/uWozY/ct/LAPbhgeEJ8PazSA+hwfDBnOXcLt0vdd4Cg==","signature_status":"signed_v1","signed_at":"2026-08-06T01:47:39.946643Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.04899","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4bfc248835499ee2e19eaae13bfeeaf6b4b1872f5dbfdfa553631238764c34eb","sha256:826bd0e2ce9efc7695a7eeb2a43a41c50c97281128fa5f416ae30ad77715bd8f"],"state_sha256":"97442745bb1121f4ad40bd215eeda23ffe1ecaf89095e62c3fb34fc897357520"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DOyFsCrvB4miqq1/9dXRfE6P9n735Ga0Es5XZK9dhNSrxavJAaqoGDCg92HdITFRh2XSahv/PbASOpA+5GfCAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T12:26:44.323240Z","bundle_sha256":"12564d3237b42ba4a7e8220d1422985d6e9336b8b9544d3c98d841092839fc31"}}