{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:YPXMCYTUK5M7YS7PDMPLAF6PU6","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":"fae2f4e3aac9bef1b343bbfee9d3c57ccaeb28d85ed11d1b9acef678726f7bdd","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2024-12-24T12:54:19Z","title_canon_sha256":"36ec16cfde4a75a9526e116bc57f4407f3e6073aeacbd1758737e0eab3d95ec7"},"schema_version":"1.0","source":{"id":"2412.18407","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.18407","created_at":"2026-07-05T11:12:23Z"},{"alias_kind":"arxiv_version","alias_value":"2412.18407v2","created_at":"2026-07-05T11:12:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.18407","created_at":"2026-07-05T11:12:23Z"},{"alias_kind":"pith_short_12","alias_value":"YPXMCYTUK5M7","created_at":"2026-07-05T11:12:23Z"},{"alias_kind":"pith_short_16","alias_value":"YPXMCYTUK5M7YS7P","created_at":"2026-07-05T11:12:23Z"},{"alias_kind":"pith_short_8","alias_value":"YPXMCYTU","created_at":"2026-07-05T11:12:23Z"}],"graph_snapshots":[{"event_id":"sha256:220decf82c57166f4453e66a931fb910dc93f6f643179784be081a7ad1bea031","target":"graph","created_at":"2026-07-05T11:12:23Z","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/2412.18407/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) have transformed natural language processing, with frameworks like Chatbot Arena providing pioneering platforms for evaluating these models. By facilitating millions of pairwise comparisons based on human judgments, Chatbot Arena has become a cornerstone in LLM evaluation, offering rich datasets for ranking models in open-ended conversational tasks. Building upon this foundation, we propose a statistical framework that incorporates key advancements to address specific challenges in pairwise comparison analysis. First, we introduce a factored tie model that enhances","authors_text":"Ion Stoica, Michael W. Mahoney, Siavash Ameli, Siyuan Zhuang","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2024-12-24T12:54:19Z","title":"A Statistical Framework for Ranking LLM-Based Chatbots"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.18407","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":"sha256:9e7b82551f700367a622b2057538460b4aeb7970970812ae0bdff10b89a701c4","target":"record","created_at":"2026-07-05T11:12:23Z","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":"fae2f4e3aac9bef1b343bbfee9d3c57ccaeb28d85ed11d1b9acef678726f7bdd","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2024-12-24T12:54:19Z","title_canon_sha256":"36ec16cfde4a75a9526e116bc57f4407f3e6073aeacbd1758737e0eab3d95ec7"},"schema_version":"1.0","source":{"id":"2412.18407","kind":"arxiv","version":2}},"canonical_sha256":"c3eec162745759fc4bef1b1eb017cfa7aace4cd834ce25f62d814b52c49620f1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c3eec162745759fc4bef1b1eb017cfa7aace4cd834ce25f62d814b52c49620f1","first_computed_at":"2026-07-05T11:12:23.785168Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:12:23.785168Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"pkmUT5ZbtHUmnAUnO7edxje9tkaNVcSp9UZdQhWr4aTwE3/WcNnLUYvzcVAVENwQtJRQ3uZPJdvl3TdeysbDCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:12:23.785697Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.18407","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9e7b82551f700367a622b2057538460b4aeb7970970812ae0bdff10b89a701c4","sha256:220decf82c57166f4453e66a931fb910dc93f6f643179784be081a7ad1bea031"],"state_sha256":"afa2609186a96ad276d7224c999dfd514890e8aa25004e24e0ccdc4332d315c3"}