{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:WGWJAW2V2GHLOJKNJBOQCSPSS4","short_pith_number":"pith:WGWJAW2V","schema_version":"1.0","canonical_sha256":"b1ac905b55d18eb7254d485d0149f2970f2dc58eab4822d5e60cf85950c46688","source":{"kind":"arxiv","id":"2602.15894","version":2},"attestation_state":"computed","paper":{"title":"Quality-constrained Entropy Maximization Policy Optimization for LLM Diversity","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Haihui Pan, Hongfei Jiang, Junwei Bao, Kaichen Zhang, Shaoke Lv, Yang Song, Yuzhong Hong","submitted_at":"2026-02-11T10:28:50Z","abstract_excerpt":"In many large language model (LLM) alignment applications, users expect not only high-quality outputs but also substantial diversity. However, existing methods often face a fundamental trade-off between these objectives: approaches that improve output quality tend to reduce diversity, while methods that increase diversity often do so at the expense of quality. In this work, we propose Quality-constrained Entropy Maximization Policy Optimization (QEMPO), a novel framework that enhances the diversity of LLM outputs while explicitly preserving output quality. QEMPO is grounded in a strong theoret"},"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":"2602.15894","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-02-11T10:28:50Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"7792a0deaa248defa6299147d9e3344a82b50b6e55e5f200eb7e130cb0dc5502","abstract_canon_sha256":"25176f1fb4fcb93c9e4507dd8058bc80101bca191d989c5c325a6dc24d35894d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:05:11.874778Z","signature_b64":"ymVCY/1E8kP38Bq91JOOLDtwxHCbo6xMrIIt0zhsZ/EzLyuob7EjuYMTDRvPir/tNJawsNr3wjGtCxop54I2Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b1ac905b55d18eb7254d485d0149f2970f2dc58eab4822d5e60cf85950c46688","last_reissued_at":"2026-05-28T01:05:11.874283Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:05:11.874283Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Quality-constrained Entropy Maximization Policy Optimization for LLM Diversity","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Haihui Pan, Hongfei Jiang, Junwei Bao, Kaichen Zhang, Shaoke Lv, Yang Song, Yuzhong Hong","submitted_at":"2026-02-11T10:28:50Z","abstract_excerpt":"In many large language model (LLM) alignment applications, users expect not only high-quality outputs but also substantial diversity. However, existing methods often face a fundamental trade-off between these objectives: approaches that improve output quality tend to reduce diversity, while methods that increase diversity often do so at the expense of quality. In this work, we propose Quality-constrained Entropy Maximization Policy Optimization (QEMPO), a novel framework that enhances the diversity of LLM outputs while explicitly preserving output quality. QEMPO is grounded in a strong theoret"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.15894","kind":"arxiv","version":2},"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/2602.15894/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":"2602.15894","created_at":"2026-05-28T01:05:11.874348+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.15894v2","created_at":"2026-05-28T01:05:11.874348+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.15894","created_at":"2026-05-28T01:05:11.874348+00:00"},{"alias_kind":"pith_short_12","alias_value":"WGWJAW2V2GHL","created_at":"2026-05-28T01:05:11.874348+00:00"},{"alias_kind":"pith_short_16","alias_value":"WGWJAW2V2GHLOJKN","created_at":"2026-05-28T01:05:11.874348+00:00"},{"alias_kind":"pith_short_8","alias_value":"WGWJAW2V","created_at":"2026-05-28T01:05:11.874348+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/WGWJAW2V2GHLOJKNJBOQCSPSS4","json":"https://pith.science/pith/WGWJAW2V2GHLOJKNJBOQCSPSS4.json","graph_json":"https://pith.science/api/pith-number/WGWJAW2V2GHLOJKNJBOQCSPSS4/graph.json","events_json":"https://pith.science/api/pith-number/WGWJAW2V2GHLOJKNJBOQCSPSS4/events.json","paper":"https://pith.science/paper/WGWJAW2V"},"agent_actions":{"view_html":"https://pith.science/pith/WGWJAW2V2GHLOJKNJBOQCSPSS4","download_json":"https://pith.science/pith/WGWJAW2V2GHLOJKNJBOQCSPSS4.json","view_paper":"https://pith.science/paper/WGWJAW2V","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.15894&json=true","fetch_graph":"https://pith.science/api/pith-number/WGWJAW2V2GHLOJKNJBOQCSPSS4/graph.json","fetch_events":"https://pith.science/api/pith-number/WGWJAW2V2GHLOJKNJBOQCSPSS4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WGWJAW2V2GHLOJKNJBOQCSPSS4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WGWJAW2V2GHLOJKNJBOQCSPSS4/action/storage_attestation","attest_author":"https://pith.science/pith/WGWJAW2V2GHLOJKNJBOQCSPSS4/action/author_attestation","sign_citation":"https://pith.science/pith/WGWJAW2V2GHLOJKNJBOQCSPSS4/action/citation_signature","submit_replication":"https://pith.science/pith/WGWJAW2V2GHLOJKNJBOQCSPSS4/action/replication_record"}},"created_at":"2026-05-28T01:05:11.874348+00:00","updated_at":"2026-05-28T01:05:11.874348+00:00"}