{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:AS4DAGOVD7LLWILH7EW3QV2F57","short_pith_number":"pith:AS4DAGOV","schema_version":"1.0","canonical_sha256":"04b83019d51fd6bb2167f92db85745efe887ff51c2757683345039ab9b841ac7","source":{"kind":"arxiv","id":"2606.09124","version":1},"attestation_state":"computed","paper":{"title":"A Regret Minimization Framework on Preference Learning in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Geon-Hyeong Kim, Jungwoo Lee, Moontae Lee, Suhwan Kim, Taehyun Cho, Youngsoo Jang, Yu Jin Kim","submitted_at":"2026-06-08T07:18:44Z","abstract_excerpt":"Reinforcement learning with verifiable rewards (RLVR) has enabled progress on reasoning-intensive tasks by relying on task-specific verifiers that provide automated correctness signals. However, many realistic language tasks are difficult to equip with reliable verifiers, motivating a growing reliance on reinforcement learning from human feedback (RLHF). In this setting, we argue that a closer examination of how human feedback should be interpreted is essential. We introduce Regret-based Preference Optimization $(\\textbf{RePO})$, which reframes RLHF through $\\textit{regret minimization}$ rathe"},"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":"2606.09124","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-08T07:18:44Z","cross_cats_sorted":[],"title_canon_sha256":"da9ce3ffd428f926e5714d5c02a4488080a4bc54f8495edb581e7a4af439b7a0","abstract_canon_sha256":"0571f006cb58d663ac072e49867a19209018109bec6b4b93720f639eb2d5f22b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T02:08:00.569569Z","signature_b64":"qOBfwVmvL9d9liqKObXZnv1z8G7/xdY4ZqD0K9q+cgylfDtoQTEbb2rFXxF2fUwAyJeZJ8PE8p0srdoG0FIYCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"04b83019d51fd6bb2167f92db85745efe887ff51c2757683345039ab9b841ac7","last_reissued_at":"2026-06-09T02:08:00.568654Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T02:08:00.568654Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Regret Minimization Framework on Preference Learning in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Geon-Hyeong Kim, Jungwoo Lee, Moontae Lee, Suhwan Kim, Taehyun Cho, Youngsoo Jang, Yu Jin Kim","submitted_at":"2026-06-08T07:18:44Z","abstract_excerpt":"Reinforcement learning with verifiable rewards (RLVR) has enabled progress on reasoning-intensive tasks by relying on task-specific verifiers that provide automated correctness signals. However, many realistic language tasks are difficult to equip with reliable verifiers, motivating a growing reliance on reinforcement learning from human feedback (RLHF). In this setting, we argue that a closer examination of how human feedback should be interpreted is essential. We introduce Regret-based Preference Optimization $(\\textbf{RePO})$, which reframes RLHF through $\\textit{regret minimization}$ rathe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.09124","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/2606.09124/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":"2606.09124","created_at":"2026-06-09T02:08:00.568804+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.09124v1","created_at":"2026-06-09T02:08:00.568804+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.09124","created_at":"2026-06-09T02:08:00.568804+00:00"},{"alias_kind":"pith_short_12","alias_value":"AS4DAGOVD7LL","created_at":"2026-06-09T02:08:00.568804+00:00"},{"alias_kind":"pith_short_16","alias_value":"AS4DAGOVD7LLWILH","created_at":"2026-06-09T02:08:00.568804+00:00"},{"alias_kind":"pith_short_8","alias_value":"AS4DAGOV","created_at":"2026-06-09T02:08:00.568804+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/AS4DAGOVD7LLWILH7EW3QV2F57","json":"https://pith.science/pith/AS4DAGOVD7LLWILH7EW3QV2F57.json","graph_json":"https://pith.science/api/pith-number/AS4DAGOVD7LLWILH7EW3QV2F57/graph.json","events_json":"https://pith.science/api/pith-number/AS4DAGOVD7LLWILH7EW3QV2F57/events.json","paper":"https://pith.science/paper/AS4DAGOV"},"agent_actions":{"view_html":"https://pith.science/pith/AS4DAGOVD7LLWILH7EW3QV2F57","download_json":"https://pith.science/pith/AS4DAGOVD7LLWILH7EW3QV2F57.json","view_paper":"https://pith.science/paper/AS4DAGOV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.09124&json=true","fetch_graph":"https://pith.science/api/pith-number/AS4DAGOVD7LLWILH7EW3QV2F57/graph.json","fetch_events":"https://pith.science/api/pith-number/AS4DAGOVD7LLWILH7EW3QV2F57/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AS4DAGOVD7LLWILH7EW3QV2F57/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AS4DAGOVD7LLWILH7EW3QV2F57/action/storage_attestation","attest_author":"https://pith.science/pith/AS4DAGOVD7LLWILH7EW3QV2F57/action/author_attestation","sign_citation":"https://pith.science/pith/AS4DAGOVD7LLWILH7EW3QV2F57/action/citation_signature","submit_replication":"https://pith.science/pith/AS4DAGOVD7LLWILH7EW3QV2F57/action/replication_record"}},"created_at":"2026-06-09T02:08:00.568804+00:00","updated_at":"2026-06-09T02:08:00.568804+00:00"}