{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:NHPKXX663VXKQ6YIWPP6NOFGMA","short_pith_number":"pith:NHPKXX66","schema_version":"1.0","canonical_sha256":"69deabdfdedd6ea87b08b3dfe6b8a66002516f944ccef5ec54d0fb5e098ffd5d","source":{"kind":"arxiv","id":"2607.10139","version":1},"attestation_state":"computed","paper":{"title":"LLMs as a Jury: Cross-Model Consensus Can Outperform Process Reward Models for LLM Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Ning Liu","submitted_at":"2026-07-11T05:57:54Z","abstract_excerpt":"Selecting the correct answer from a pool of candidate reasoning chains is the engine of test-time scaling, yet the standard selectors each carry a cost: self-consistency inherits the errors of the single model it resamples, and trained reward models need labeled data and transfer poorly off-distribution. We study a third signal, free at inference time: cross-model consensus, the degree to which independently trained models, each solving the problem once, agree on a final answer. We treat the panel as an LLM-jury, in which the structure of agreement, not any model's score of another, is the ver"},"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":"2607.10139","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-11T05:57:54Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3806aa82b3a71fbebaefd31924da653d033bb3edaf8a4b0d6c9b0b0c28f6bb94","abstract_canon_sha256":"ba1225a6e8065f6e363bcb23282f1775282f7227103dec34011f343bc2b3fca1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-14T01:20:27.237169Z","signature_b64":"0lhA+8iIptUsm359uuNttaEi5l2LROCQtuQZjw+mphPh3omzDt05BPUbbhs/Fa5o6eEUnv+XuiSYuLXvhPOZCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"69deabdfdedd6ea87b08b3dfe6b8a66002516f944ccef5ec54d0fb5e098ffd5d","last_reissued_at":"2026-07-14T01:20:27.236388Z","signature_status":"signed_v1","first_computed_at":"2026-07-14T01:20:27.236388Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LLMs as a Jury: Cross-Model Consensus Can Outperform Process Reward Models for LLM Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Ning Liu","submitted_at":"2026-07-11T05:57:54Z","abstract_excerpt":"Selecting the correct answer from a pool of candidate reasoning chains is the engine of test-time scaling, yet the standard selectors each carry a cost: self-consistency inherits the errors of the single model it resamples, and trained reward models need labeled data and transfer poorly off-distribution. We study a third signal, free at inference time: cross-model consensus, the degree to which independently trained models, each solving the problem once, agree on a final answer. We treat the panel as an LLM-jury, in which the structure of agreement, not any model's score of another, is the ver"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.10139","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/2607.10139/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":"2607.10139","created_at":"2026-07-14T01:20:27.236793+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.10139v1","created_at":"2026-07-14T01:20:27.236793+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.10139","created_at":"2026-07-14T01:20:27.236793+00:00"},{"alias_kind":"pith_short_12","alias_value":"NHPKXX663VXK","created_at":"2026-07-14T01:20:27.236793+00:00"},{"alias_kind":"pith_short_16","alias_value":"NHPKXX663VXKQ6YI","created_at":"2026-07-14T01:20:27.236793+00:00"},{"alias_kind":"pith_short_8","alias_value":"NHPKXX66","created_at":"2026-07-14T01:20:27.236793+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/NHPKXX663VXKQ6YIWPP6NOFGMA","json":"https://pith.science/pith/NHPKXX663VXKQ6YIWPP6NOFGMA.json","graph_json":"https://pith.science/api/pith-number/NHPKXX663VXKQ6YIWPP6NOFGMA/graph.json","events_json":"https://pith.science/api/pith-number/NHPKXX663VXKQ6YIWPP6NOFGMA/events.json","paper":"https://pith.science/paper/NHPKXX66"},"agent_actions":{"view_html":"https://pith.science/pith/NHPKXX663VXKQ6YIWPP6NOFGMA","download_json":"https://pith.science/pith/NHPKXX663VXKQ6YIWPP6NOFGMA.json","view_paper":"https://pith.science/paper/NHPKXX66","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.10139&json=true","fetch_graph":"https://pith.science/api/pith-number/NHPKXX663VXKQ6YIWPP6NOFGMA/graph.json","fetch_events":"https://pith.science/api/pith-number/NHPKXX663VXKQ6YIWPP6NOFGMA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NHPKXX663VXKQ6YIWPP6NOFGMA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NHPKXX663VXKQ6YIWPP6NOFGMA/action/storage_attestation","attest_author":"https://pith.science/pith/NHPKXX663VXKQ6YIWPP6NOFGMA/action/author_attestation","sign_citation":"https://pith.science/pith/NHPKXX663VXKQ6YIWPP6NOFGMA/action/citation_signature","submit_replication":"https://pith.science/pith/NHPKXX663VXKQ6YIWPP6NOFGMA/action/replication_record"}},"created_at":"2026-07-14T01:20:27.236793+00:00","updated_at":"2026-07-14T01:20:27.236793+00:00"}