{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:32LOQJ22AIKQ4CKTBV455QVTA5","short_pith_number":"pith:32LOQJ22","schema_version":"1.0","canonical_sha256":"de96e8275a02150e09530d79dec2b30745bdaef69a559ba66808aed5491ad153","source":{"kind":"arxiv","id":"2605.30758","version":1},"attestation_state":"computed","paper":{"title":"Pairwise Reference Alignment as a Model-Level Ordinal Observable","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Mujing Li","submitted_at":"2026-05-29T02:41:18Z","abstract_excerpt":"Pairwise preference data is widely used in language-model evaluation and alignment, often for model ranking, reward modeling, or preference optimization. This note formulates a more basic measurement question: given a reference distribution of pairwise preferences, what model-level quantity is estimated when we test whether a model ranks preferred responses above rejected responses?\n  We define pairwise reference alignment as an ordinal observable induced by a model scoring function. Given a reference pair distribution $P_{\\mathrm{pair}}$ over triples $(x,y^+,y^-)$, and a scalar model score $S"},"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":"2605.30758","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-29T02:41:18Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"3a4468bfcb0606b53a3c28fe6db3161771998c58f494e1ffabf95a4943de74bb","abstract_canon_sha256":"5aded4f58b6282275c3f56cddf6f5fc3640c1c80994a0b7fff12556cd6bba431"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T01:03:14.671543Z","signature_b64":"0yFlUCnGGAHgB1DzmHi4QNjL4fGkfLOMnwmga3MuibtTh2BTeO1vbE/jMJJu5MQVfMfgpM40BQd6VumhMjzkCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"de96e8275a02150e09530d79dec2b30745bdaef69a559ba66808aed5491ad153","last_reissued_at":"2026-06-01T01:03:14.670816Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T01:03:14.670816Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Pairwise Reference Alignment as a Model-Level Ordinal Observable","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Mujing Li","submitted_at":"2026-05-29T02:41:18Z","abstract_excerpt":"Pairwise preference data is widely used in language-model evaluation and alignment, often for model ranking, reward modeling, or preference optimization. This note formulates a more basic measurement question: given a reference distribution of pairwise preferences, what model-level quantity is estimated when we test whether a model ranks preferred responses above rejected responses?\n  We define pairwise reference alignment as an ordinal observable induced by a model scoring function. Given a reference pair distribution $P_{\\mathrm{pair}}$ over triples $(x,y^+,y^-)$, and a scalar model score $S"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.30758","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/2605.30758/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":"2605.30758","created_at":"2026-06-01T01:03:14.670934+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.30758v1","created_at":"2026-06-01T01:03:14.670934+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.30758","created_at":"2026-06-01T01:03:14.670934+00:00"},{"alias_kind":"pith_short_12","alias_value":"32LOQJ22AIKQ","created_at":"2026-06-01T01:03:14.670934+00:00"},{"alias_kind":"pith_short_16","alias_value":"32LOQJ22AIKQ4CKT","created_at":"2026-06-01T01:03:14.670934+00:00"},{"alias_kind":"pith_short_8","alias_value":"32LOQJ22","created_at":"2026-06-01T01:03:14.670934+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/32LOQJ22AIKQ4CKTBV455QVTA5","json":"https://pith.science/pith/32LOQJ22AIKQ4CKTBV455QVTA5.json","graph_json":"https://pith.science/api/pith-number/32LOQJ22AIKQ4CKTBV455QVTA5/graph.json","events_json":"https://pith.science/api/pith-number/32LOQJ22AIKQ4CKTBV455QVTA5/events.json","paper":"https://pith.science/paper/32LOQJ22"},"agent_actions":{"view_html":"https://pith.science/pith/32LOQJ22AIKQ4CKTBV455QVTA5","download_json":"https://pith.science/pith/32LOQJ22AIKQ4CKTBV455QVTA5.json","view_paper":"https://pith.science/paper/32LOQJ22","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.30758&json=true","fetch_graph":"https://pith.science/api/pith-number/32LOQJ22AIKQ4CKTBV455QVTA5/graph.json","fetch_events":"https://pith.science/api/pith-number/32LOQJ22AIKQ4CKTBV455QVTA5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/32LOQJ22AIKQ4CKTBV455QVTA5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/32LOQJ22AIKQ4CKTBV455QVTA5/action/storage_attestation","attest_author":"https://pith.science/pith/32LOQJ22AIKQ4CKTBV455QVTA5/action/author_attestation","sign_citation":"https://pith.science/pith/32LOQJ22AIKQ4CKTBV455QVTA5/action/citation_signature","submit_replication":"https://pith.science/pith/32LOQJ22AIKQ4CKTBV455QVTA5/action/replication_record"}},"created_at":"2026-06-01T01:03:14.670934+00:00","updated_at":"2026-06-01T01:03:14.670934+00:00"}