{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:SNLIWIGDZBCGMAGDRNZSIH6YAV","short_pith_number":"pith:SNLIWIGD","schema_version":"1.0","canonical_sha256":"93568b20c3c8446600c38b73241fd80564de0fd0b8ffeac2c3902f8b5d671ec5","source":{"kind":"arxiv","id":"2606.20820","version":1},"attestation_state":"computed","paper":{"title":"CELEUS: Certifiable and Efficient LLM Evaluation via E-Processes","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bo Li, Feng Liu, Zesheng Ye, Zhaorun Chen, Zhijian Zhou","submitted_at":"2026-06-18T18:07:42Z","abstract_excerpt":"Can we trust evaluation scores to capture an LLM's true real-world performance? Certifiable evaluation answers this question by providing guarantee for LLM evaluation. In particular, existing methods sequentially curate evaluation samples and keep updating confidence intervals (CIs) that cover the true performance with high probability (e.g., 95%) until some conditions are satisfied, e.g., the CI width reaches a target precision. However, existing methods are not generally anytime-valid: the claimed coverage (e.g., 95%) may fail when CIs are repeatedly updated and used to decide when to stop, "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.20820","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-18T18:07:42Z","cross_cats_sorted":[],"title_canon_sha256":"64cb398bed768345e6e596161bfb17be1b295e908992f009b6d582fc961f9d3b","abstract_canon_sha256":"658c417f90042ef47329dea58cb60344aabee7ba0c43b3b8010af6bca0a4b0eb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T00:11:59.704835Z","signature_b64":"0YfMnQBkepGkESlfaAMGv/m1fUlR7etIbq58+3sMAWW6XqxEgpvfvR5W8hiEXJrXHkDj4EaPaBLVyYMRVdT9Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"93568b20c3c8446600c38b73241fd80564de0fd0b8ffeac2c3902f8b5d671ec5","last_reissued_at":"2026-06-23T00:11:59.704447Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T00:11:59.704447Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CELEUS: Certifiable and Efficient LLM Evaluation via E-Processes","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bo Li, Feng Liu, Zesheng Ye, Zhaorun Chen, Zhijian Zhou","submitted_at":"2026-06-18T18:07:42Z","abstract_excerpt":"Can we trust evaluation scores to capture an LLM's true real-world performance? Certifiable evaluation answers this question by providing guarantee for LLM evaluation. In particular, existing methods sequentially curate evaluation samples and keep updating confidence intervals (CIs) that cover the true performance with high probability (e.g., 95%) until some conditions are satisfied, e.g., the CI width reaches a target precision. However, existing methods are not generally anytime-valid: the claimed coverage (e.g., 95%) may fail when CIs are repeatedly updated and used to decide when to stop, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.20820","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/2606.20820/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.20820","created_at":"2026-06-23T00:11:59.704503+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.20820v1","created_at":"2026-06-23T00:11:59.704503+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.20820","created_at":"2026-06-23T00:11:59.704503+00:00"},{"alias_kind":"pith_short_12","alias_value":"SNLIWIGDZBCG","created_at":"2026-06-23T00:11:59.704503+00:00"},{"alias_kind":"pith_short_16","alias_value":"SNLIWIGDZBCGMAGD","created_at":"2026-06-23T00:11:59.704503+00:00"},{"alias_kind":"pith_short_8","alias_value":"SNLIWIGD","created_at":"2026-06-23T00:11:59.704503+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SNLIWIGDZBCGMAGDRNZSIH6YAV","json":"https://pith.science/pith/SNLIWIGDZBCGMAGDRNZSIH6YAV.json","graph_json":"https://pith.science/api/pith-number/SNLIWIGDZBCGMAGDRNZSIH6YAV/graph.json","events_json":"https://pith.science/api/pith-number/SNLIWIGDZBCGMAGDRNZSIH6YAV/events.json","paper":"https://pith.science/paper/SNLIWIGD"},"agent_actions":{"view_html":"https://pith.science/pith/SNLIWIGDZBCGMAGDRNZSIH6YAV","download_json":"https://pith.science/pith/SNLIWIGDZBCGMAGDRNZSIH6YAV.json","view_paper":"https://pith.science/paper/SNLIWIGD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.20820&json=true","fetch_graph":"https://pith.science/api/pith-number/SNLIWIGDZBCGMAGDRNZSIH6YAV/graph.json","fetch_events":"https://pith.science/api/pith-number/SNLIWIGDZBCGMAGDRNZSIH6YAV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SNLIWIGDZBCGMAGDRNZSIH6YAV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SNLIWIGDZBCGMAGDRNZSIH6YAV/action/storage_attestation","attest_author":"https://pith.science/pith/SNLIWIGDZBCGMAGDRNZSIH6YAV/action/author_attestation","sign_citation":"https://pith.science/pith/SNLIWIGDZBCGMAGDRNZSIH6YAV/action/citation_signature","submit_replication":"https://pith.science/pith/SNLIWIGDZBCGMAGDRNZSIH6YAV/action/replication_record"}},"created_at":"2026-06-23T00:11:59.704503+00:00","updated_at":"2026-06-23T00:11:59.704503+00:00"}