{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:V67NSWBDXGIYCGTNUF4ITXQ7PE","short_pith_number":"pith:V67NSWBD","schema_version":"1.0","canonical_sha256":"afbed95823b991811a6da17889de1f792a5b315f33516f996c8d3267482e54d9","source":{"kind":"arxiv","id":"2606.10078","version":1},"attestation_state":"computed","paper":{"title":"Mult-DPO: Multinomial Direct Preference Optimization for Recommender Systems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Aditya Sinha, Harald Steck, James McInerney, Jundong Li, Nathan Kallus, Yaochen Zhu, Yinhan He","submitted_at":"2026-06-08T18:53:45Z","abstract_excerpt":"Direct preference optimization (DPO) is a simple and effective alignment strategy for large language models (LLMs) based on pairwise preferences. In recommender systems, however, user feedback is rarely pairwise. For a given context, e.g., a user, a session, or a conversation, we typically observe set-wise preferences with multiple positive items, where every positive item should outrank every unobserved or explicitly negative item, with no prescribed order among the positives or the negatives themselves. A natural generalization is to use the Plackett-Luce (PL) reward model, which extends the"},"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":"2606.10078","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2026-06-08T18:53:45Z","cross_cats_sorted":[],"title_canon_sha256":"99dd149ffb05d8038055edd5e89a53346171b4182a61d3d877b6c4e0d6f74fc7","abstract_canon_sha256":"6bb0254893bb6ed8f45b6653d451359086a4c167b8ae3649b73951c5843fbda6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-10T00:08:48.228222Z","signature_b64":"0GNiPHgzC6KD81AllvlKS6ufntT6mf5uBelo84Nja4RO5rWZXR8IueiEyoq+6lTHTmFWbFRchVuzUnvPspyjDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"afbed95823b991811a6da17889de1f792a5b315f33516f996c8d3267482e54d9","last_reissued_at":"2026-06-10T00:08:48.227327Z","signature_status":"signed_v1","first_computed_at":"2026-06-10T00:08:48.227327Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Mult-DPO: Multinomial Direct Preference Optimization for Recommender Systems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Aditya Sinha, Harald Steck, James McInerney, Jundong Li, Nathan Kallus, Yaochen Zhu, Yinhan He","submitted_at":"2026-06-08T18:53:45Z","abstract_excerpt":"Direct preference optimization (DPO) is a simple and effective alignment strategy for large language models (LLMs) based on pairwise preferences. In recommender systems, however, user feedback is rarely pairwise. For a given context, e.g., a user, a session, or a conversation, we typically observe set-wise preferences with multiple positive items, where every positive item should outrank every unobserved or explicitly negative item, with no prescribed order among the positives or the negatives themselves. A natural generalization is to use the Plackett-Luce (PL) reward model, which extends the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.10078","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/2606.10078/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":"2606.10078","created_at":"2026-06-10T00:08:48.227476+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.10078v1","created_at":"2026-06-10T00:08:48.227476+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.10078","created_at":"2026-06-10T00:08:48.227476+00:00"},{"alias_kind":"pith_short_12","alias_value":"V67NSWBDXGIY","created_at":"2026-06-10T00:08:48.227476+00:00"},{"alias_kind":"pith_short_16","alias_value":"V67NSWBDXGIYCGTN","created_at":"2026-06-10T00:08:48.227476+00:00"},{"alias_kind":"pith_short_8","alias_value":"V67NSWBD","created_at":"2026-06-10T00:08:48.227476+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/V67NSWBDXGIYCGTNUF4ITXQ7PE","json":"https://pith.science/pith/V67NSWBDXGIYCGTNUF4ITXQ7PE.json","graph_json":"https://pith.science/api/pith-number/V67NSWBDXGIYCGTNUF4ITXQ7PE/graph.json","events_json":"https://pith.science/api/pith-number/V67NSWBDXGIYCGTNUF4ITXQ7PE/events.json","paper":"https://pith.science/paper/V67NSWBD"},"agent_actions":{"view_html":"https://pith.science/pith/V67NSWBDXGIYCGTNUF4ITXQ7PE","download_json":"https://pith.science/pith/V67NSWBDXGIYCGTNUF4ITXQ7PE.json","view_paper":"https://pith.science/paper/V67NSWBD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.10078&json=true","fetch_graph":"https://pith.science/api/pith-number/V67NSWBDXGIYCGTNUF4ITXQ7PE/graph.json","fetch_events":"https://pith.science/api/pith-number/V67NSWBDXGIYCGTNUF4ITXQ7PE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V67NSWBDXGIYCGTNUF4ITXQ7PE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V67NSWBDXGIYCGTNUF4ITXQ7PE/action/storage_attestation","attest_author":"https://pith.science/pith/V67NSWBDXGIYCGTNUF4ITXQ7PE/action/author_attestation","sign_citation":"https://pith.science/pith/V67NSWBDXGIYCGTNUF4ITXQ7PE/action/citation_signature","submit_replication":"https://pith.science/pith/V67NSWBDXGIYCGTNUF4ITXQ7PE/action/replication_record"}},"created_at":"2026-06-10T00:08:48.227476+00:00","updated_at":"2026-06-10T00:08:48.227476+00:00"}