{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:KVEJ77XWX3OK736IFD4F3PAXDR","short_pith_number":"pith:KVEJ77XW","canonical_record":{"source":{"id":"2401.12794","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-23T14:29:17Z","cross_cats_sorted":[],"title_canon_sha256":"44b96fa7ce0bdf51123806868b7423b8c716abf4806a2ea084474fd43843777a","abstract_canon_sha256":"a37929bb11a3397da1245a1b2a38d5d8808cea820b1bb31ebcb30e64498f9e7b"},"schema_version":"1.0"},"canonical_sha256":"55489ffef6bedcafefc828f85dbc171c5124273fda923e097aa09df2df2e4423","source":{"kind":"arxiv","id":"2401.12794","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.12794","created_at":"2026-07-05T09:29:13Z"},{"alias_kind":"arxiv_version","alias_value":"2401.12794v3","created_at":"2026-07-05T09:29:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.12794","created_at":"2026-07-05T09:29:13Z"},{"alias_kind":"pith_short_12","alias_value":"KVEJ77XWX3OK","created_at":"2026-07-05T09:29:13Z"},{"alias_kind":"pith_short_16","alias_value":"KVEJ77XWX3OK736I","created_at":"2026-07-05T09:29:13Z"},{"alias_kind":"pith_short_8","alias_value":"KVEJ77XW","created_at":"2026-07-05T09:29:13Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:KVEJ77XWX3OK736IFD4F3PAXDR","target":"record","payload":{"canonical_record":{"source":{"id":"2401.12794","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-23T14:29:17Z","cross_cats_sorted":[],"title_canon_sha256":"44b96fa7ce0bdf51123806868b7423b8c716abf4806a2ea084474fd43843777a","abstract_canon_sha256":"a37929bb11a3397da1245a1b2a38d5d8808cea820b1bb31ebcb30e64498f9e7b"},"schema_version":"1.0"},"canonical_sha256":"55489ffef6bedcafefc828f85dbc171c5124273fda923e097aa09df2df2e4423","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:29:13.728685Z","signature_b64":"R1/SKGiD0Ro19ZaLj5zQu9sFraFqRUFqFgHPZT780FBGUk8ZqFROMRw8agwCstjsms31uSQ7noFtxtubuyiyBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"55489ffef6bedcafefc828f85dbc171c5124273fda923e097aa09df2df2e4423","last_reissued_at":"2026-07-05T09:29:13.728193Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:29:13.728193Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2401.12794","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:29:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IYeAhCYYaVROAS7ZPatfHdLVSOdlP5EZ3EXBiCdkxclDYH2XVSWu42hoxqaXZT5TfSA6c+G4GvyJ7Oyc96coBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T07:21:59.077858Z"},"content_sha256":"35ad03ea09576590e39ac2f450d0fe1c344601801c47b58b6f3204686b6b5ab6","schema_version":"1.0","event_id":"sha256:35ad03ea09576590e39ac2f450d0fe1c344601801c47b58b6f3204686b6b5ab6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:KVEJ77XWX3OK736IFD4F3PAXDR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Benchmarking LLMs via Uncertainty Quantification","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Derek F. Wong, Emine Yilmaz, Fanghua Ye, Jianhui Pang, Longyue Wang, Mingming Yang, Shuming Shi, Zhaopeng Tu","submitted_at":"2024-01-23T14:29:17Z","abstract_excerpt":"The proliferation of open-source Large Language Models (LLMs) from various institutions has highlighted the urgent need for comprehensive evaluation methods. However, current evaluation platforms, such as the widely recognized HuggingFace open LLM leaderboard, neglect a crucial aspect -- uncertainty, which is vital for thoroughly assessing LLMs. To bridge this gap, we introduce a new benchmarking approach for LLMs that integrates uncertainty quantification. Our examination involves nine LLMs (LLM series) spanning five representative natural language processing tasks. Our findings reveal that: "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.12794","kind":"arxiv","version":3},"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/2401.12794/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":"2026-07-05T09:29:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gk0jT/5YVemLLyvCExd6YsI1xEzOwQenJqCWn3Fe/pfJMOGbIbXWYH7LVYLgqNqvuCy/YpBfangvbc32NFXABQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T07:21:59.078734Z"},"content_sha256":"fb226bc69bf678d2a74998303d5718af31bdb8f0b35ee0dab618ab4fbce25727","schema_version":"1.0","event_id":"sha256:fb226bc69bf678d2a74998303d5718af31bdb8f0b35ee0dab618ab4fbce25727"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KVEJ77XWX3OK736IFD4F3PAXDR/bundle.json","state_url":"https://pith.science/pith/KVEJ77XWX3OK736IFD4F3PAXDR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KVEJ77XWX3OK736IFD4F3PAXDR/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":"2026-08-08T07:21:59Z","links":{"resolver":"https://pith.science/pith/KVEJ77XWX3OK736IFD4F3PAXDR","bundle":"https://pith.science/pith/KVEJ77XWX3OK736IFD4F3PAXDR/bundle.json","state":"https://pith.science/pith/KVEJ77XWX3OK736IFD4F3PAXDR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KVEJ77XWX3OK736IFD4F3PAXDR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:KVEJ77XWX3OK736IFD4F3PAXDR","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":"a37929bb11a3397da1245a1b2a38d5d8808cea820b1bb31ebcb30e64498f9e7b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-23T14:29:17Z","title_canon_sha256":"44b96fa7ce0bdf51123806868b7423b8c716abf4806a2ea084474fd43843777a"},"schema_version":"1.0","source":{"id":"2401.12794","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.12794","created_at":"2026-07-05T09:29:13Z"},{"alias_kind":"arxiv_version","alias_value":"2401.12794v3","created_at":"2026-07-05T09:29:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.12794","created_at":"2026-07-05T09:29:13Z"},{"alias_kind":"pith_short_12","alias_value":"KVEJ77XWX3OK","created_at":"2026-07-05T09:29:13Z"},{"alias_kind":"pith_short_16","alias_value":"KVEJ77XWX3OK736I","created_at":"2026-07-05T09:29:13Z"},{"alias_kind":"pith_short_8","alias_value":"KVEJ77XW","created_at":"2026-07-05T09:29:13Z"}],"graph_snapshots":[{"event_id":"sha256:fb226bc69bf678d2a74998303d5718af31bdb8f0b35ee0dab618ab4fbce25727","target":"graph","created_at":"2026-07-05T09:29:13Z","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/2401.12794/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The proliferation of open-source Large Language Models (LLMs) from various institutions has highlighted the urgent need for comprehensive evaluation methods. However, current evaluation platforms, such as the widely recognized HuggingFace open LLM leaderboard, neglect a crucial aspect -- uncertainty, which is vital for thoroughly assessing LLMs. To bridge this gap, we introduce a new benchmarking approach for LLMs that integrates uncertainty quantification. Our examination involves nine LLMs (LLM series) spanning five representative natural language processing tasks. Our findings reveal that: ","authors_text":"Derek F. Wong, Emine Yilmaz, Fanghua Ye, Jianhui Pang, Longyue Wang, Mingming Yang, Shuming Shi, Zhaopeng Tu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-23T14:29:17Z","title":"Benchmarking LLMs via Uncertainty Quantification"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.12794","kind":"arxiv","version":3},"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":"sha256:35ad03ea09576590e39ac2f450d0fe1c344601801c47b58b6f3204686b6b5ab6","target":"record","created_at":"2026-07-05T09:29:13Z","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":"a37929bb11a3397da1245a1b2a38d5d8808cea820b1bb31ebcb30e64498f9e7b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-23T14:29:17Z","title_canon_sha256":"44b96fa7ce0bdf51123806868b7423b8c716abf4806a2ea084474fd43843777a"},"schema_version":"1.0","source":{"id":"2401.12794","kind":"arxiv","version":3}},"canonical_sha256":"55489ffef6bedcafefc828f85dbc171c5124273fda923e097aa09df2df2e4423","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"55489ffef6bedcafefc828f85dbc171c5124273fda923e097aa09df2df2e4423","first_computed_at":"2026-07-05T09:29:13.728193Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:29:13.728193Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"R1/SKGiD0Ro19ZaLj5zQu9sFraFqRUFqFgHPZT780FBGUk8ZqFROMRw8agwCstjsms31uSQ7noFtxtubuyiyBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:29:13.728685Z","signed_message":"canonical_sha256_bytes"},"source_id":"2401.12794","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:35ad03ea09576590e39ac2f450d0fe1c344601801c47b58b6f3204686b6b5ab6","sha256:fb226bc69bf678d2a74998303d5718af31bdb8f0b35ee0dab618ab4fbce25727"],"state_sha256":"ab4797cbc0d9e1bb85558f97ec7924f6b8f05a85142c6e72959dd3fbb725626f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IGTxFWRP9M+JYO+h06B4UenT7Nc+g0u71ZfVlyzLpaFr+HslahCWoar+nBZSboewH1+KjuoRNSvKUc9y1nhSBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T07:21:59.084316Z","bundle_sha256":"492cc8b7958451b4117e540d8373643ea6554b62a090b9f7690e7379abcbd9e3"}}