{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:HKG6ZIF6M62J4TFERT5V7GPAHI","short_pith_number":"pith:HKG6ZIF6","canonical_record":{"source":{"id":"2402.03282","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-05T18:38:55Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"b381542f1279c16429d2c5882092d501fdee609005badc6837290f45dc8df1dd","abstract_canon_sha256":"970e1734c6b792b0535d8b1ba55b99a9b30c560b692305e8ddee2b2898e9d509"},"schema_version":"1.0"},"canonical_sha256":"3a8deca0be67b49e4ca48cfb5f99e03a0b4fd840164a7b78a626b631f8746b41","source":{"kind":"arxiv","id":"2402.03282","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.03282","created_at":"2026-07-05T09:33:08Z"},{"alias_kind":"arxiv_version","alias_value":"2402.03282v3","created_at":"2026-07-05T09:33:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.03282","created_at":"2026-07-05T09:33:08Z"},{"alias_kind":"pith_short_12","alias_value":"HKG6ZIF6M62J","created_at":"2026-07-05T09:33:08Z"},{"alias_kind":"pith_short_16","alias_value":"HKG6ZIF6M62J4TFE","created_at":"2026-07-05T09:33:08Z"},{"alias_kind":"pith_short_8","alias_value":"HKG6ZIF6","created_at":"2026-07-05T09:33:08Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:HKG6ZIF6M62J4TFERT5V7GPAHI","target":"record","payload":{"canonical_record":{"source":{"id":"2402.03282","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-05T18:38:55Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"b381542f1279c16429d2c5882092d501fdee609005badc6837290f45dc8df1dd","abstract_canon_sha256":"970e1734c6b792b0535d8b1ba55b99a9b30c560b692305e8ddee2b2898e9d509"},"schema_version":"1.0"},"canonical_sha256":"3a8deca0be67b49e4ca48cfb5f99e03a0b4fd840164a7b78a626b631f8746b41","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:33:08.079236Z","signature_b64":"9oWkQgywwS3QJ4iwRLQvUOkOyxyb5HbaOk2KEdOjA/yIJASN0kf+kb/67udTkLJyqz1rKiOwqVzWF3/l1GPcDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3a8deca0be67b49e4ca48cfb5f99e03a0b4fd840164a7b78a626b631f8746b41","last_reissued_at":"2026-07-05T09:33:08.078630Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:33:08.078630Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.03282","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:33:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vjL1VL+f7hoszQsrvJv4HFFaqCbOlXAPgZOPsQz7ZfUGtsu0xDb87NmDVGGVvEpuYl+/ttzdYJHgkjJBmklqDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T16:50:29.660406Z"},"content_sha256":"4974d2cf34918883fc183a6b7a466c8b0039e65053f7b7c5aea9d46beb4bef35","schema_version":"1.0","event_id":"sha256:4974d2cf34918883fc183a6b7a466c8b0039e65053f7b7c5aea9d46beb4bef35"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:HKG6ZIF6M62J4TFERT5V7GPAHI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A Theoretical Framework for Partially Observed Reward-States in RLHF","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Aldo Pacchiano, Ambuj Tewari, Chinmaya Kausik, Mirco Mutti","submitted_at":"2024-02-05T18:38:55Z","abstract_excerpt":"The growing deployment of reinforcement learning from human feedback (RLHF) calls for a deeper theoretical investigation of its underlying models. The prevalent models of RLHF do not account for neuroscience-backed, partially-observed \"internal states\" that can affect human feedback, nor do they accommodate intermediate feedback during an interaction. Both of these can be instrumental in speeding up learning and improving alignment. To address these limitations, we model RLHF as reinforcement learning with partially observed reward-states (PORRL). We accommodate two kinds of feedback $-$ cardi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.03282","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/2402.03282/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":"2026-07-05T09:33:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"k0T4JyK3X3V6f1zYavcg2WCjMS5N4KO9qA3GFGm2DlO3rIUCNbx0VjMkgljnzKosz1xbKl3h/fxy9CmIshvDBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T16:50:29.660955Z"},"content_sha256":"8e3dd8f57e5d32b2438ffb24949c424d1aa7eb11138ee14dc89b1b2c4bfb2018","schema_version":"1.0","event_id":"sha256:8e3dd8f57e5d32b2438ffb24949c424d1aa7eb11138ee14dc89b1b2c4bfb2018"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HKG6ZIF6M62J4TFERT5V7GPAHI/bundle.json","state_url":"https://pith.science/pith/HKG6ZIF6M62J4TFERT5V7GPAHI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HKG6ZIF6M62J4TFERT5V7GPAHI/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":"2026-08-10T16:50:29Z","links":{"resolver":"https://pith.science/pith/HKG6ZIF6M62J4TFERT5V7GPAHI","bundle":"https://pith.science/pith/HKG6ZIF6M62J4TFERT5V7GPAHI/bundle.json","state":"https://pith.science/pith/HKG6ZIF6M62J4TFERT5V7GPAHI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HKG6ZIF6M62J4TFERT5V7GPAHI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:HKG6ZIF6M62J4TFERT5V7GPAHI","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":"970e1734c6b792b0535d8b1ba55b99a9b30c560b692305e8ddee2b2898e9d509","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-05T18:38:55Z","title_canon_sha256":"b381542f1279c16429d2c5882092d501fdee609005badc6837290f45dc8df1dd"},"schema_version":"1.0","source":{"id":"2402.03282","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.03282","created_at":"2026-07-05T09:33:08Z"},{"alias_kind":"arxiv_version","alias_value":"2402.03282v3","created_at":"2026-07-05T09:33:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.03282","created_at":"2026-07-05T09:33:08Z"},{"alias_kind":"pith_short_12","alias_value":"HKG6ZIF6M62J","created_at":"2026-07-05T09:33:08Z"},{"alias_kind":"pith_short_16","alias_value":"HKG6ZIF6M62J4TFE","created_at":"2026-07-05T09:33:08Z"},{"alias_kind":"pith_short_8","alias_value":"HKG6ZIF6","created_at":"2026-07-05T09:33:08Z"}],"graph_snapshots":[{"event_id":"sha256:8e3dd8f57e5d32b2438ffb24949c424d1aa7eb11138ee14dc89b1b2c4bfb2018","target":"graph","created_at":"2026-07-05T09:33:08Z","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/2402.03282/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The growing deployment of reinforcement learning from human feedback (RLHF) calls for a deeper theoretical investigation of its underlying models. The prevalent models of RLHF do not account for neuroscience-backed, partially-observed \"internal states\" that can affect human feedback, nor do they accommodate intermediate feedback during an interaction. Both of these can be instrumental in speeding up learning and improving alignment. To address these limitations, we model RLHF as reinforcement learning with partially observed reward-states (PORRL). We accommodate two kinds of feedback $-$ cardi","authors_text":"Aldo Pacchiano, Ambuj Tewari, Chinmaya Kausik, Mirco Mutti","cross_cats":["cs.AI","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-05T18:38:55Z","title":"A Theoretical Framework for Partially Observed Reward-States in RLHF"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.03282","kind":"arxiv","version":3},"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":"sha256:4974d2cf34918883fc183a6b7a466c8b0039e65053f7b7c5aea9d46beb4bef35","target":"record","created_at":"2026-07-05T09:33:08Z","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":"970e1734c6b792b0535d8b1ba55b99a9b30c560b692305e8ddee2b2898e9d509","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-05T18:38:55Z","title_canon_sha256":"b381542f1279c16429d2c5882092d501fdee609005badc6837290f45dc8df1dd"},"schema_version":"1.0","source":{"id":"2402.03282","kind":"arxiv","version":3}},"canonical_sha256":"3a8deca0be67b49e4ca48cfb5f99e03a0b4fd840164a7b78a626b631f8746b41","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3a8deca0be67b49e4ca48cfb5f99e03a0b4fd840164a7b78a626b631f8746b41","first_computed_at":"2026-07-05T09:33:08.078630Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:33:08.078630Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9oWkQgywwS3QJ4iwRLQvUOkOyxyb5HbaOk2KEdOjA/yIJASN0kf+kb/67udTkLJyqz1rKiOwqVzWF3/l1GPcDg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:33:08.079236Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.03282","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4974d2cf34918883fc183a6b7a466c8b0039e65053f7b7c5aea9d46beb4bef35","sha256:8e3dd8f57e5d32b2438ffb24949c424d1aa7eb11138ee14dc89b1b2c4bfb2018"],"state_sha256":"f0f3c47267fba8b31cb090c451b42756f0472555bedf44827a467a7f57448861"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"euMxzUAaRoGiIWK8FNjBBOJ4wdy1WStBL5BSqOmqcXy6k1nuRN0K+yMay5LhNUCVJNzg5unMjn/2BYUShJE7Cg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T16:50:29.665611Z","bundle_sha256":"c73aa03fbde8a6aad17f164573829f800156ccfbd2658ab08c88f608fea6b0fe"}}