{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:CVKUROCCVEMGDBCA6XXC3AYTNW","short_pith_number":"pith:CVKUROCC","schema_version":"1.0","canonical_sha256":"155548b842a918618440f5ee2d83136d99b19e18245dd34de5ee171fa44ffded","source":{"kind":"arxiv","id":"2601.21817","version":2},"attestation_state":"computed","paper":{"title":"A Judge-Aware Ranking Framework for Evaluating Large Language Models without Ground Truth","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Doudou Zhou, Jiawei Wu, Mingyuan Xu, Xinzi Tan","submitted_at":"2026-01-29T15:01:28Z","abstract_excerpt":"Evaluating large language models (LLMs) on open-ended tasks without ground-truth labels is increasingly done via the LLM-as-a-judge paradigm. A critical but under-modeled issue is that judge LLMs differ substantially in reliability; treating all judges equally can yield biased leaderboards and misleading uncertainty estimates. More data can make evaluation more confidently wrong under misspecified aggregation. We propose a judge-aware ranking framework that extends the Bradley-Terry-Luce model by introducing judge-specific discrimination parameters, jointly estimating latent model quality and "},"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":"2601.21817","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2026-01-29T15:01:28Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"44d5c5f4c594fe4b37f68815cd8ccc8b2b55a51688adb71083b5fe2bfcb9abfb","abstract_canon_sha256":"2c9f4624d135604100504c581f551077d8f024cd94f48b6f7266bf160a3458d3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-05T01:14:33.824922Z","signature_b64":"jNxV8cL6MZLAbFJrduHCs0VHcJvlDvBU1JjpHKhhi4umjFo3AjwgDvL003seBZe2Yr8SWujX98p56YwPVSx+AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"155548b842a918618440f5ee2d83136d99b19e18245dd34de5ee171fa44ffded","last_reissued_at":"2026-06-05T01:14:33.824115Z","signature_status":"signed_v1","first_computed_at":"2026-06-05T01:14:33.824115Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Judge-Aware Ranking Framework for Evaluating Large Language Models without Ground Truth","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Doudou Zhou, Jiawei Wu, Mingyuan Xu, Xinzi Tan","submitted_at":"2026-01-29T15:01:28Z","abstract_excerpt":"Evaluating large language models (LLMs) on open-ended tasks without ground-truth labels is increasingly done via the LLM-as-a-judge paradigm. A critical but under-modeled issue is that judge LLMs differ substantially in reliability; treating all judges equally can yield biased leaderboards and misleading uncertainty estimates. More data can make evaluation more confidently wrong under misspecified aggregation. We propose a judge-aware ranking framework that extends the Bradley-Terry-Luce model by introducing judge-specific discrimination parameters, jointly estimating latent model quality and "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.21817","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/2601.21817/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":"2601.21817","created_at":"2026-06-05T01:14:33.824210+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.21817v2","created_at":"2026-06-05T01:14:33.824210+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.21817","created_at":"2026-06-05T01:14:33.824210+00:00"},{"alias_kind":"pith_short_12","alias_value":"CVKUROCCVEMG","created_at":"2026-06-05T01:14:33.824210+00:00"},{"alias_kind":"pith_short_16","alias_value":"CVKUROCCVEMGDBCA","created_at":"2026-06-05T01:14:33.824210+00:00"},{"alias_kind":"pith_short_8","alias_value":"CVKUROCC","created_at":"2026-06-05T01:14:33.824210+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2605.09702","citing_title":"Calibrate, Don't Curate: Label-Efficient Estimation from Noisy LLM Judges","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2605.05073","citing_title":"Heterogeneous Judge-Aware Ranking with Sensitivity, Disagreement, and Confidence","ref_index":17,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CVKUROCCVEMGDBCA6XXC3AYTNW","json":"https://pith.science/pith/CVKUROCCVEMGDBCA6XXC3AYTNW.json","graph_json":"https://pith.science/api/pith-number/CVKUROCCVEMGDBCA6XXC3AYTNW/graph.json","events_json":"https://pith.science/api/pith-number/CVKUROCCVEMGDBCA6XXC3AYTNW/events.json","paper":"https://pith.science/paper/CVKUROCC"},"agent_actions":{"view_html":"https://pith.science/pith/CVKUROCCVEMGDBCA6XXC3AYTNW","download_json":"https://pith.science/pith/CVKUROCCVEMGDBCA6XXC3AYTNW.json","view_paper":"https://pith.science/paper/CVKUROCC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.21817&json=true","fetch_graph":"https://pith.science/api/pith-number/CVKUROCCVEMGDBCA6XXC3AYTNW/graph.json","fetch_events":"https://pith.science/api/pith-number/CVKUROCCVEMGDBCA6XXC3AYTNW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CVKUROCCVEMGDBCA6XXC3AYTNW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CVKUROCCVEMGDBCA6XXC3AYTNW/action/storage_attestation","attest_author":"https://pith.science/pith/CVKUROCCVEMGDBCA6XXC3AYTNW/action/author_attestation","sign_citation":"https://pith.science/pith/CVKUROCCVEMGDBCA6XXC3AYTNW/action/citation_signature","submit_replication":"https://pith.science/pith/CVKUROCCVEMGDBCA6XXC3AYTNW/action/replication_record"}},"created_at":"2026-06-05T01:14:33.824210+00:00","updated_at":"2026-06-05T01:14:33.824210+00:00"}