{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:OCDA5YZYQG7C4GCMWEHIQYIKBL","short_pith_number":"pith:OCDA5YZY","schema_version":"1.0","canonical_sha256":"70860ee33881be2e184cb10e88610a0accf138faeff3940d0f0e08da73186f67","source":{"kind":"arxiv","id":"2308.02585","version":3},"attestation_state":"computed","paper":{"title":"PARL: A Unified Framework for Policy Alignment in Reinforcement Learning from Human Feedback","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alec Koppel, Amrit Singh Bedi, Dinesh Manocha, Furong Huang, Huazheng Wang, Mengdi Wang, Souradip Chakraborty","submitted_at":"2023-08-03T18:03:44Z","abstract_excerpt":"We present a novel unified bilevel optimization-based framework, \\textsf{PARL}, formulated to address the recently highlighted critical issue of policy alignment in reinforcement learning using utility or preference-based feedback. We identify a major gap within current algorithmic designs for solving policy alignment due to a lack of precise characterization of the dependence of the alignment objective on the data generated by policy trajectories. This shortfall contributes to the sub-optimal performance observed in contemporary algorithms. Our framework addressed these concerns by explicitly"},"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":"2308.02585","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-08-03T18:03:44Z","cross_cats_sorted":[],"title_canon_sha256":"192717f906a4b1843513f7eeb8505d8c2accdb93f644be0a2d07d855f6fb648d","abstract_canon_sha256":"93083cf0b479c1efd051627d62c35c36f57e2f6381bbdbe19aae3463cb519e06"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:14:00.392080Z","signature_b64":"OP0wXOcQGhUxTsm7/27LFDNmOUfARuCUV8S8rMVDLdp80wCM5Md8MntaNGwEb06ZTSZe5Oy//laCvvZLxVwxBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"70860ee33881be2e184cb10e88610a0accf138faeff3940d0f0e08da73186f67","last_reissued_at":"2026-07-05T08:14:00.391567Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:14:00.391567Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PARL: A Unified Framework for Policy Alignment in Reinforcement Learning from Human Feedback","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alec Koppel, Amrit Singh Bedi, Dinesh Manocha, Furong Huang, Huazheng Wang, Mengdi Wang, Souradip Chakraborty","submitted_at":"2023-08-03T18:03:44Z","abstract_excerpt":"We present a novel unified bilevel optimization-based framework, \\textsf{PARL}, formulated to address the recently highlighted critical issue of policy alignment in reinforcement learning using utility or preference-based feedback. We identify a major gap within current algorithmic designs for solving policy alignment due to a lack of precise characterization of the dependence of the alignment objective on the data generated by policy trajectories. This shortfall contributes to the sub-optimal performance observed in contemporary algorithms. Our framework addressed these concerns by explicitly"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.02585","kind":"arxiv","version":3},"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/2308.02585/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":"2308.02585","created_at":"2026-07-05T08:14:00.391627+00:00"},{"alias_kind":"arxiv_version","alias_value":"2308.02585v3","created_at":"2026-07-05T08:14:00.391627+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.02585","created_at":"2026-07-05T08:14:00.391627+00:00"},{"alias_kind":"pith_short_12","alias_value":"OCDA5YZYQG7C","created_at":"2026-07-05T08:14:00.391627+00:00"},{"alias_kind":"pith_short_16","alias_value":"OCDA5YZYQG7C4GCM","created_at":"2026-07-05T08:14:00.391627+00:00"},{"alias_kind":"pith_short_8","alias_value":"OCDA5YZY","created_at":"2026-07-05T08:14:00.391627+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/OCDA5YZYQG7C4GCMWEHIQYIKBL","json":"https://pith.science/pith/OCDA5YZYQG7C4GCMWEHIQYIKBL.json","graph_json":"https://pith.science/api/pith-number/OCDA5YZYQG7C4GCMWEHIQYIKBL/graph.json","events_json":"https://pith.science/api/pith-number/OCDA5YZYQG7C4GCMWEHIQYIKBL/events.json","paper":"https://pith.science/paper/OCDA5YZY"},"agent_actions":{"view_html":"https://pith.science/pith/OCDA5YZYQG7C4GCMWEHIQYIKBL","download_json":"https://pith.science/pith/OCDA5YZYQG7C4GCMWEHIQYIKBL.json","view_paper":"https://pith.science/paper/OCDA5YZY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2308.02585&json=true","fetch_graph":"https://pith.science/api/pith-number/OCDA5YZYQG7C4GCMWEHIQYIKBL/graph.json","fetch_events":"https://pith.science/api/pith-number/OCDA5YZYQG7C4GCMWEHIQYIKBL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OCDA5YZYQG7C4GCMWEHIQYIKBL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OCDA5YZYQG7C4GCMWEHIQYIKBL/action/storage_attestation","attest_author":"https://pith.science/pith/OCDA5YZYQG7C4GCMWEHIQYIKBL/action/author_attestation","sign_citation":"https://pith.science/pith/OCDA5YZYQG7C4GCMWEHIQYIKBL/action/citation_signature","submit_replication":"https://pith.science/pith/OCDA5YZYQG7C4GCMWEHIQYIKBL/action/replication_record"}},"created_at":"2026-07-05T08:14:00.391627+00:00","updated_at":"2026-07-05T08:14:00.391627+00:00"}