{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:7QRY2QYBWBEKFDH2GGXCLVGRIQ","short_pith_number":"pith:7QRY2QYB","schema_version":"1.0","canonical_sha256":"fc238d4301b048a28cfa31ae25d4d14423a53f4db7ba5ac288705b1423f600f6","source":{"kind":"arxiv","id":"2607.27422","version":1},"attestation_state":"computed","paper":{"title":"Good Rankers, Bad Objectives: Bilinear Contrastive Critics under Expressive Policy Search","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ayushman Singh, Siddharth Aphale","submitted_at":"2026-07-29T19:43:40Z","abstract_excerpt":"Good action rankings do not make a contrastive critic safe to maximize. These critics increasingly act as value-like objectives for best-of-$K$ selection, planning, and critic-guided generation. Unbounded bilinear scores can let large embedding norms inflate off-support values, but cosine bounding does not remove the failure. A controlled support decomposition attributes most raw bilinear regret to norm drift. Cosine and hybrid critics nevertheless select off-support actions from most pools and incur comparable regret. Contrastive scores are weakly calibrated or inverted in the top score decil"},"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":"2607.27422","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-07-29T19:43:40Z","cross_cats_sorted":[],"title_canon_sha256":"4fe21787177a1e7fe82e2dafe1d8039cf96ab06884632b69d7b9b31a25fac524","abstract_canon_sha256":"0f2b456716138158e950b3d1663f682eeca177ffc4949f53d4fc924738578092"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fc238d4301b048a28cfa31ae25d4d14423a53f4db7ba5ac288705b1423f600f6","last_reissued_at":"2026-07-31T00:11:59.706174Z","signature_status":"unsigned_v0","first_computed_at":"2026-07-31T00:11:59.706174Z"},"graph_snapshot":{"paper":{"title":"Good Rankers, Bad Objectives: Bilinear Contrastive Critics under Expressive Policy Search","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ayushman Singh, Siddharth Aphale","submitted_at":"2026-07-29T19:43:40Z","abstract_excerpt":"Good action rankings do not make a contrastive critic safe to maximize. These critics increasingly act as value-like objectives for best-of-$K$ selection, planning, and critic-guided generation. Unbounded bilinear scores can let large embedding norms inflate off-support values, but cosine bounding does not remove the failure. A controlled support decomposition attributes most raw bilinear regret to norm drift. Cosine and hybrid critics nevertheless select off-support actions from most pools and incur comparable regret. Contrastive scores are weakly calibrated or inverted in the top score decil"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.27422","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/2607.27422/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":"2607.27422","created_at":"2026-07-31T00:11:59.708519+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.27422v1","created_at":"2026-07-31T00:11:59.708519+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.27422","created_at":"2026-07-31T00:11:59.708519+00:00"},{"alias_kind":"pith_short_12","alias_value":"7QRY2QYBWBEK","created_at":"2026-07-31T00:11:59.708519+00:00"},{"alias_kind":"pith_short_16","alias_value":"7QRY2QYBWBEKFDH2","created_at":"2026-07-31T00:11:59.708519+00:00"},{"alias_kind":"pith_short_8","alias_value":"7QRY2QYB","created_at":"2026-07-31T00:11:59.708519+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/7QRY2QYBWBEKFDH2GGXCLVGRIQ","json":"https://pith.science/pith/7QRY2QYBWBEKFDH2GGXCLVGRIQ.json","graph_json":"https://pith.science/api/pith-number/7QRY2QYBWBEKFDH2GGXCLVGRIQ/graph.json","events_json":"https://pith.science/api/pith-number/7QRY2QYBWBEKFDH2GGXCLVGRIQ/events.json","paper":"https://pith.science/paper/7QRY2QYB"},"agent_actions":{"view_html":"https://pith.science/pith/7QRY2QYBWBEKFDH2GGXCLVGRIQ","download_json":"https://pith.science/pith/7QRY2QYBWBEKFDH2GGXCLVGRIQ.json","view_paper":"https://pith.science/paper/7QRY2QYB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.27422&json=true","fetch_graph":"https://pith.science/api/pith-number/7QRY2QYBWBEKFDH2GGXCLVGRIQ/graph.json","fetch_events":"https://pith.science/api/pith-number/7QRY2QYBWBEKFDH2GGXCLVGRIQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7QRY2QYBWBEKFDH2GGXCLVGRIQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7QRY2QYBWBEKFDH2GGXCLVGRIQ/action/storage_attestation","attest_author":"https://pith.science/pith/7QRY2QYBWBEKFDH2GGXCLVGRIQ/action/author_attestation","sign_citation":"https://pith.science/pith/7QRY2QYBWBEKFDH2GGXCLVGRIQ/action/citation_signature","submit_replication":"https://pith.science/pith/7QRY2QYBWBEKFDH2GGXCLVGRIQ/action/replication_record"}},"created_at":"2026-07-31T00:11:59.708519+00:00","updated_at":"2026-07-31T00:11:59.708519+00:00"}