{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:XYYEOBJLJUVNNQ2LKMNEGY5HBO","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":"1ac437431f22e93a477288c30498adb43f39ae0ca79fd66792599a22e5067f67","cross_cats_sorted":["stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ME","submitted_at":"2026-05-28T05:44:43Z","title_canon_sha256":"50f0a515284e7aadb81ffca265c4807648ec0cd65b99dcec5087ac3e41ba2705"},"schema_version":"1.0","source":{"id":"2605.29395","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.29395","created_at":"2026-05-29T01:05:36Z"},{"alias_kind":"arxiv_version","alias_value":"2605.29395v1","created_at":"2026-05-29T01:05:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.29395","created_at":"2026-05-29T01:05:36Z"},{"alias_kind":"pith_short_12","alias_value":"XYYEOBJLJUVN","created_at":"2026-05-29T01:05:36Z"},{"alias_kind":"pith_short_16","alias_value":"XYYEOBJLJUVNNQ2L","created_at":"2026-05-29T01:05:36Z"},{"alias_kind":"pith_short_8","alias_value":"XYYEOBJL","created_at":"2026-05-29T01:05:36Z"}],"graph_snapshots":[{"event_id":"sha256:abeee72d6760dd7f751d2f1c05b6e503f476688e02ccc6976d912b20020decef","target":"graph","created_at":"2026-05-29T01:05:36Z","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/2605.29395/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Pairwise human-preference platforms such as Chatbot Arena have become central to large language model (LLM) evaluation, yet reliable task-specific ranking remains challenging. Global leaderboards mask task heterogeneity, while ranking each fine-grained task independently is unstable under sparse, imbalanced comparisons. We propose a low-rank framework for task-specific LLM ranking from sparse pairwise comparisons, modeling the task-by-model ability matrix $\\Theta^\\star \\in \\mathbb{R}^{d_t \\times d_m}$ as low rank so that information is shared across related tasks while task-specific difference","authors_text":"David Simchi-Levi, Jiachun Li, Will Wei Sun","cross_cats":["stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ME","submitted_at":"2026-05-28T05:44:43Z","title":"Low Rank for Rank: Uncertainty-Aware Task-Specific LLM Ranking under Sparse Pairwise Comparisons"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.29395","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":"sha256:28ba64bf771230e3b2168c3f455b0ca3e3ae4ca7a8909d5c2b8a31d5a2fc6f5c","target":"record","created_at":"2026-05-29T01:05:36Z","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":"1ac437431f22e93a477288c30498adb43f39ae0ca79fd66792599a22e5067f67","cross_cats_sorted":["stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ME","submitted_at":"2026-05-28T05:44:43Z","title_canon_sha256":"50f0a515284e7aadb81ffca265c4807648ec0cd65b99dcec5087ac3e41ba2705"},"schema_version":"1.0","source":{"id":"2605.29395","kind":"arxiv","version":1}},"canonical_sha256":"be3047052b4d2ad6c34b531a4363a70bb0abdee0015a6062dd6aff2f936a94c9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"be3047052b4d2ad6c34b531a4363a70bb0abdee0015a6062dd6aff2f936a94c9","first_computed_at":"2026-05-29T01:05:36.839620Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-29T01:05:36.839620Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"C6kpRWDUCvs18j+nsqRiEtzwvsaKqZAYFoZ5Mr8pfIM6L4KwIIcxSusbwjLP1FGzdxaiZsRghDFHY/GrTas7AQ==","signature_status":"signed_v1","signed_at":"2026-05-29T01:05:36.840428Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.29395","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:28ba64bf771230e3b2168c3f455b0ca3e3ae4ca7a8909d5c2b8a31d5a2fc6f5c","sha256:abeee72d6760dd7f751d2f1c05b6e503f476688e02ccc6976d912b20020decef"],"state_sha256":"e60f0ed26d23df9299b790830d668a3ac4ff2b6cb97643d1beefbdd7bbaffb03"}