{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:BX53GY7KEQTTPCBDJUCFAAFPK2","short_pith_number":"pith:BX53GY7K","schema_version":"1.0","canonical_sha256":"0dfbb363ea24273788234d045000af568b87370ae78f30fffdedefd0a30fcc10","source":{"kind":"arxiv","id":"2608.00175","version":1},"attestation_state":"computed","paper":{"title":"Inference-Time Policy Alignment for Fair Reinforcement Learning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Conor Wallace, Peilang Li, Umer Siddique, Yongcan Cao","submitted_at":"2026-07-31T18:00:50Z","abstract_excerpt":"Deep reinforcement learning (RL) agents achieve strong performance by optimizing scalar reward functions. However, once deployed, the policies of these RL agents are often rigid and costly to adapt to new performance criteria. For instance, an agent trained to maximize expected cumulative reward may not accommodate previously unknown stakeholder preferences. Existing approaches to achieve fairness, a type of preference, in RL typically assume that such preferences are known a priori and require complete retraining of the policy under a fairness-oriented metric. Inspired by inference-time align"},"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":"2608.00175","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-31T18:00:50Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"91e73b774f0530d7723dca4f3b462d53915f371d6a1d71867c97d3045047f014","abstract_canon_sha256":"36d5f960088839d15ad3e96229c7eb0f723f452c1616a41bb595788ceb179b16"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-04T00:33:11.167842Z","signature_b64":"mIxn4Ozc0Aqy9r8RpfoFSr0vR2LSMoiEpMf0FEHEpLItUcrHSBU4aFKHtWdwraxCQG4WYs+DEbESIJaiaTidAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0dfbb363ea24273788234d045000af568b87370ae78f30fffdedefd0a30fcc10","last_reissued_at":"2026-08-04T00:33:11.166278Z","signature_status":"signed_v1","first_computed_at":"2026-08-04T00:33:11.166278Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Inference-Time Policy Alignment for Fair Reinforcement Learning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Conor Wallace, Peilang Li, Umer Siddique, Yongcan Cao","submitted_at":"2026-07-31T18:00:50Z","abstract_excerpt":"Deep reinforcement learning (RL) agents achieve strong performance by optimizing scalar reward functions. However, once deployed, the policies of these RL agents are often rigid and costly to adapt to new performance criteria. For instance, an agent trained to maximize expected cumulative reward may not accommodate previously unknown stakeholder preferences. Existing approaches to achieve fairness, a type of preference, in RL typically assume that such preferences are known a priori and require complete retraining of the policy under a fairness-oriented metric. Inspired by inference-time align"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.00175","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/2608.00175/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":"2608.00175","created_at":"2026-08-04T00:33:11.167406+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.00175v1","created_at":"2026-08-04T00:33:11.167406+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.00175","created_at":"2026-08-04T00:33:11.167406+00:00"},{"alias_kind":"pith_short_12","alias_value":"BX53GY7KEQTT","created_at":"2026-08-04T00:33:11.167406+00:00"},{"alias_kind":"pith_short_16","alias_value":"BX53GY7KEQTTPCBD","created_at":"2026-08-04T00:33:11.167406+00:00"},{"alias_kind":"pith_short_8","alias_value":"BX53GY7K","created_at":"2026-08-04T00:33:11.167406+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/BX53GY7KEQTTPCBDJUCFAAFPK2","json":"https://pith.science/pith/BX53GY7KEQTTPCBDJUCFAAFPK2.json","graph_json":"https://pith.science/api/pith-number/BX53GY7KEQTTPCBDJUCFAAFPK2/graph.json","events_json":"https://pith.science/api/pith-number/BX53GY7KEQTTPCBDJUCFAAFPK2/events.json","paper":"https://pith.science/paper/BX53GY7K"},"agent_actions":{"view_html":"https://pith.science/pith/BX53GY7KEQTTPCBDJUCFAAFPK2","download_json":"https://pith.science/pith/BX53GY7KEQTTPCBDJUCFAAFPK2.json","view_paper":"https://pith.science/paper/BX53GY7K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.00175&json=true","fetch_graph":"https://pith.science/api/pith-number/BX53GY7KEQTTPCBDJUCFAAFPK2/graph.json","fetch_events":"https://pith.science/api/pith-number/BX53GY7KEQTTPCBDJUCFAAFPK2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BX53GY7KEQTTPCBDJUCFAAFPK2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BX53GY7KEQTTPCBDJUCFAAFPK2/action/storage_attestation","attest_author":"https://pith.science/pith/BX53GY7KEQTTPCBDJUCFAAFPK2/action/author_attestation","sign_citation":"https://pith.science/pith/BX53GY7KEQTTPCBDJUCFAAFPK2/action/citation_signature","submit_replication":"https://pith.science/pith/BX53GY7KEQTTPCBDJUCFAAFPK2/action/replication_record"}},"created_at":"2026-08-04T00:33:11.167406+00:00","updated_at":"2026-08-04T00:33:11.167406+00:00"}