{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:G7SFOWGLITV6ZHFWUQ5O37SC5L","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":"f036b071636c10c1482546ced597bab06e8f9a9418f0453aed0bcff7503dc8a5","cross_cats_sorted":["cs.AI","econ.EM","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-12-26T08:22:41Z","title_canon_sha256":"d3ff80c6183a697a842c1de14e4f4d2f498a0852e25e1b263398bc0a6f55c090"},"schema_version":"1.0","source":{"id":"2512.21917","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2512.21917","created_at":"2026-06-04T01:09:40Z"},{"alias_kind":"arxiv_version","alias_value":"2512.21917v3","created_at":"2026-06-04T01:09:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.21917","created_at":"2026-06-04T01:09:40Z"},{"alias_kind":"pith_short_12","alias_value":"G7SFOWGLITV6","created_at":"2026-06-04T01:09:40Z"},{"alias_kind":"pith_short_16","alias_value":"G7SFOWGLITV6ZHFW","created_at":"2026-06-04T01:09:40Z"},{"alias_kind":"pith_short_8","alias_value":"G7SFOWGL","created_at":"2026-06-04T01:09:40Z"}],"graph_snapshots":[{"event_id":"sha256:0cb9427cf66d69d31ea926fa1666974518aa2c95372e66d9ccaa5b911abec143","target":"graph","created_at":"2026-06-04T01:09:40Z","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/2512.21917/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Policy alignment to preference data typically assumes a known link function between observed preferences and latent rewards (e.g., Bradley-Terry model / logistic link). Misspecification of this link can bias inferred rewards and misalign learned policies. We study policy alignment under an unknown and unrestricted link function. We formulate an $f$-divergence-constrained reward maximization problem and show that realizability in a policy class induces a semiparametric single-index binary choice model, where a scalar policy-induced index captures all dependence on demonstrations and the remaini","authors_text":"Nathan Kallus","cross_cats":["cs.AI","econ.EM","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-12-26T08:22:41Z","title":"Semiparametric Preference Optimization: Your Language Model is Secretly a Single-Index Model"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2512.21917","kind":"arxiv","version":3},"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":"sha256:5becaa063585a6a19b9742c411a0ed18ee84d838b15b517ffa9d045e3d259276","target":"record","created_at":"2026-06-04T01:09:40Z","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":"f036b071636c10c1482546ced597bab06e8f9a9418f0453aed0bcff7503dc8a5","cross_cats_sorted":["cs.AI","econ.EM","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-12-26T08:22:41Z","title_canon_sha256":"d3ff80c6183a697a842c1de14e4f4d2f498a0852e25e1b263398bc0a6f55c090"},"schema_version":"1.0","source":{"id":"2512.21917","kind":"arxiv","version":3}},"canonical_sha256":"37e45758cb44ebec9cb6a43aedfe42eadb88e72780b3bdff93ad1b062aca605c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"37e45758cb44ebec9cb6a43aedfe42eadb88e72780b3bdff93ad1b062aca605c","first_computed_at":"2026-06-04T01:09:40.566623Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-04T01:09:40.566623Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"76U/dDA992NoMppoO2k+69d9NtO4Yf71c1as0NtDoUXiPzh5EQDCpFDC+H0QKE5ZJiTVRTCauvhiQsRVN41ICQ==","signature_status":"signed_v1","signed_at":"2026-06-04T01:09:40.567137Z","signed_message":"canonical_sha256_bytes"},"source_id":"2512.21917","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5becaa063585a6a19b9742c411a0ed18ee84d838b15b517ffa9d045e3d259276","sha256:0cb9427cf66d69d31ea926fa1666974518aa2c95372e66d9ccaa5b911abec143"],"state_sha256":"eece932135681df284e6ab07042310bc60f4bb10fa075e27ec9f6931d12a67ee"}