{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:PJQUY7FF727GAO43CNYNEK7QJ5","short_pith_number":"pith:PJQUY7FF","canonical_record":{"source":{"id":"2506.09183","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-10T19:00:19Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"31be6190a82f2926c6e8d1324ae29f6ef993c542dc2ed82e7a978c61ef799c89","abstract_canon_sha256":"03120a48b59cb1b2e575ddffbd9f356181a6a3ec25aa9ef239a9b4da28419994"},"schema_version":"1.0"},"canonical_sha256":"7a614c7ca5febe603b9b1370d22bf04f4b145110731dbde21556c23f20860d50","source":{"kind":"arxiv","id":"2506.09183","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.09183","created_at":"2026-07-05T11:23:29Z"},{"alias_kind":"arxiv_version","alias_value":"2506.09183v2","created_at":"2026-07-05T11:23:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.09183","created_at":"2026-07-05T11:23:29Z"},{"alias_kind":"pith_short_12","alias_value":"PJQUY7FF727G","created_at":"2026-07-05T11:23:29Z"},{"alias_kind":"pith_short_16","alias_value":"PJQUY7FF727GAO43","created_at":"2026-07-05T11:23:29Z"},{"alias_kind":"pith_short_8","alias_value":"PJQUY7FF","created_at":"2026-07-05T11:23:29Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:PJQUY7FF727GAO43CNYNEK7QJ5","target":"record","payload":{"canonical_record":{"source":{"id":"2506.09183","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-10T19:00:19Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"31be6190a82f2926c6e8d1324ae29f6ef993c542dc2ed82e7a978c61ef799c89","abstract_canon_sha256":"03120a48b59cb1b2e575ddffbd9f356181a6a3ec25aa9ef239a9b4da28419994"},"schema_version":"1.0"},"canonical_sha256":"7a614c7ca5febe603b9b1370d22bf04f4b145110731dbde21556c23f20860d50","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:23:29.721350Z","signature_b64":"1hrRQ0L7xILv28kYPu3rxNhJdIc46E1ZA0j+ms63A7oLCqNaVWohsThcJTiPwyvxDBhON0kR8ZwgfNHmgL5aAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7a614c7ca5febe603b9b1370d22bf04f4b145110731dbde21556c23f20860d50","last_reissued_at":"2026-07-05T11:23:29.719738Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:23:29.719738Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.09183","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:23:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"W+6xn2JW7c1BEDt6s0ze21aPetGdSihPPFrfghYrJ/pyUDOBFGVnwbr2hQVW3LIufVfHOYfhrS/hYI//3GxfCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T23:16:00.076071Z"},"content_sha256":"0ca8cc502e22f825bb89b34ac0055ad20d9e7c03ee0ccf896063c50dbef23ac4","schema_version":"1.0","event_id":"sha256:0ca8cc502e22f825bb89b34ac0055ad20d9e7c03ee0ccf896063c50dbef23ac4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:PJQUY7FF727GAO43CNYNEK7QJ5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Multi-Task Reward Learning from Human Ratings","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Devin White, Evelyn Rose, Mingkang Wu, Nicholas R Waytowich, Vernon Lawhern, Yongcan Cao","submitted_at":"2025-06-10T19:00:19Z","abstract_excerpt":"Reinforcement learning from human feedback (RLHF) has become a key factor in aligning model behavior with users' goals. However, while humans integrate multiple strategies when making decisions, current RLHF approaches often simplify this process by modeling human reasoning through isolated tasks such as classification or regression. In this paper, we propose a novel reinforcement learning (RL) method that mimics human decision-making by jointly considering multiple tasks. Specifically, we leverage human ratings in reward-free environments to infer a reward function, introducing learnable weig"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.09183","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/2506.09183/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:23:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"h07VZT1OCfSptA6Q7chO5x4eym1Yav2tEVnLLax3xcoyhkE2dA2O9U5B8BEJJVuE37O6pc6Khz0ZgZ3OpBNIBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T23:16:00.076793Z"},"content_sha256":"8126cd6369c1ee204fef8de136acdc4897c67e6e5fca783196538b2fed8e453d","schema_version":"1.0","event_id":"sha256:8126cd6369c1ee204fef8de136acdc4897c67e6e5fca783196538b2fed8e453d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PJQUY7FF727GAO43CNYNEK7QJ5/bundle.json","state_url":"https://pith.science/pith/PJQUY7FF727GAO43CNYNEK7QJ5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PJQUY7FF727GAO43CNYNEK7QJ5/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-09T23:16:00Z","links":{"resolver":"https://pith.science/pith/PJQUY7FF727GAO43CNYNEK7QJ5","bundle":"https://pith.science/pith/PJQUY7FF727GAO43CNYNEK7QJ5/bundle.json","state":"https://pith.science/pith/PJQUY7FF727GAO43CNYNEK7QJ5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PJQUY7FF727GAO43CNYNEK7QJ5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:PJQUY7FF727GAO43CNYNEK7QJ5","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"03120a48b59cb1b2e575ddffbd9f356181a6a3ec25aa9ef239a9b4da28419994","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-10T19:00:19Z","title_canon_sha256":"31be6190a82f2926c6e8d1324ae29f6ef993c542dc2ed82e7a978c61ef799c89"},"schema_version":"1.0","source":{"id":"2506.09183","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.09183","created_at":"2026-07-05T11:23:29Z"},{"alias_kind":"arxiv_version","alias_value":"2506.09183v2","created_at":"2026-07-05T11:23:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.09183","created_at":"2026-07-05T11:23:29Z"},{"alias_kind":"pith_short_12","alias_value":"PJQUY7FF727G","created_at":"2026-07-05T11:23:29Z"},{"alias_kind":"pith_short_16","alias_value":"PJQUY7FF727GAO43","created_at":"2026-07-05T11:23:29Z"},{"alias_kind":"pith_short_8","alias_value":"PJQUY7FF","created_at":"2026-07-05T11:23:29Z"}],"graph_snapshots":[{"event_id":"sha256:8126cd6369c1ee204fef8de136acdc4897c67e6e5fca783196538b2fed8e453d","target":"graph","created_at":"2026-07-05T11:23:29Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2506.09183/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning from human feedback (RLHF) has become a key factor in aligning model behavior with users' goals. However, while humans integrate multiple strategies when making decisions, current RLHF approaches often simplify this process by modeling human reasoning through isolated tasks such as classification or regression. In this paper, we propose a novel reinforcement learning (RL) method that mimics human decision-making by jointly considering multiple tasks. Specifically, we leverage human ratings in reward-free environments to infer a reward function, introducing learnable weig","authors_text":"Devin White, Evelyn Rose, Mingkang Wu, Nicholas R Waytowich, Vernon Lawhern, Yongcan Cao","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-10T19:00:19Z","title":"Multi-Task Reward Learning from Human Ratings"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.09183","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:0ca8cc502e22f825bb89b34ac0055ad20d9e7c03ee0ccf896063c50dbef23ac4","target":"record","created_at":"2026-07-05T11:23:29Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"03120a48b59cb1b2e575ddffbd9f356181a6a3ec25aa9ef239a9b4da28419994","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-10T19:00:19Z","title_canon_sha256":"31be6190a82f2926c6e8d1324ae29f6ef993c542dc2ed82e7a978c61ef799c89"},"schema_version":"1.0","source":{"id":"2506.09183","kind":"arxiv","version":2}},"canonical_sha256":"7a614c7ca5febe603b9b1370d22bf04f4b145110731dbde21556c23f20860d50","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7a614c7ca5febe603b9b1370d22bf04f4b145110731dbde21556c23f20860d50","first_computed_at":"2026-07-05T11:23:29.719738Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:23:29.719738Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1hrRQ0L7xILv28kYPu3rxNhJdIc46E1ZA0j+ms63A7oLCqNaVWohsThcJTiPwyvxDBhON0kR8ZwgfNHmgL5aAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:23:29.721350Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.09183","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0ca8cc502e22f825bb89b34ac0055ad20d9e7c03ee0ccf896063c50dbef23ac4","sha256:8126cd6369c1ee204fef8de136acdc4897c67e6e5fca783196538b2fed8e453d"],"state_sha256":"0a1dce683e0e970286c6df100a79d031f4364223816b2e0eac7c3c54b10cce5e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zolRWvzR/Y0L3E0mj2/Qpypes6UL9P50veNCBJWs0IlP2QjsfNB5S6BzAl73uvOSvpptYJVCwbRUhDG5JqW1Dw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T23:16:00.081297Z","bundle_sha256":"9ed603d3c82e7bd581f93eef6cda13a3060c7e8f93bfb500cf13c35e418f5809"}}