{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:LGLZBBSUS26O6CMRN7KSPLUTQN","short_pith_number":"pith:LGLZBBSU","canonical_record":{"source":{"id":"2402.00782","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-01T17:10:35Z","cross_cats_sorted":[],"title_canon_sha256":"2163f390390ba469ea48dc253268f9d39b104ad277229e9583fcfd6feff9c179","abstract_canon_sha256":"f46cf919706d875f4480a757a35e23fa7b95da7e7cfda3053abede483922f4ba"},"schema_version":"1.0"},"canonical_sha256":"599790865496bcef09916fd527ae93834a13dd979fa74d0381c661ec02bd9943","source":{"kind":"arxiv","id":"2402.00782","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.00782","created_at":"2026-07-05T07:40:17Z"},{"alias_kind":"arxiv_version","alias_value":"2402.00782v1","created_at":"2026-07-05T07:40:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.00782","created_at":"2026-07-05T07:40:17Z"},{"alias_kind":"pith_short_12","alias_value":"LGLZBBSUS26O","created_at":"2026-07-05T07:40:17Z"},{"alias_kind":"pith_short_16","alias_value":"LGLZBBSUS26O6CMR","created_at":"2026-07-05T07:40:17Z"},{"alias_kind":"pith_short_8","alias_value":"LGLZBBSU","created_at":"2026-07-05T07:40:17Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:LGLZBBSUS26O6CMRN7KSPLUTQN","target":"record","payload":{"canonical_record":{"source":{"id":"2402.00782","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-01T17:10:35Z","cross_cats_sorted":[],"title_canon_sha256":"2163f390390ba469ea48dc253268f9d39b104ad277229e9583fcfd6feff9c179","abstract_canon_sha256":"f46cf919706d875f4480a757a35e23fa7b95da7e7cfda3053abede483922f4ba"},"schema_version":"1.0"},"canonical_sha256":"599790865496bcef09916fd527ae93834a13dd979fa74d0381c661ec02bd9943","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:40:17.617871Z","signature_b64":"6ykQwH2NvhXpXBSeQHpAcljSnQXGeSbrHaqnJkF4MU0Atkqp2hwl9Qp+FGi9rgnxvje5XPcN07+Z4OGr5E3CDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"599790865496bcef09916fd527ae93834a13dd979fa74d0381c661ec02bd9943","last_reissued_at":"2026-07-05T07:40:17.617458Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:40:17.617458Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.00782","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:40:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Bs0EkZkWoet9QPtD4uCM5Hp63n0FIO0dAbTbKXXH/JTgyXOP+6knp3lEkk0ZrkN1MRHxgN7TLd3czivkRx6xDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T10:53:33.192901Z"},"content_sha256":"24bb7b1291f696b93574f71365a5c285c06524b7e835f9aeb154ca77a4851f6e","schema_version":"1.0","event_id":"sha256:24bb7b1291f696b93574f71365a5c285c06524b7e835f9aeb154ca77a4851f6e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:LGLZBBSUS26O6CMRN7KSPLUTQN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Dense Reward for Free in Reinforcement Learning from Human Feedback","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alex J. Chan, Hao Sun, Mihaela van der Schaar, Samuel Holt","submitted_at":"2024-02-01T17:10:35Z","abstract_excerpt":"Reinforcement Learning from Human Feedback (RLHF) has been credited as the key advance that has allowed Large Language Models (LLMs) to effectively follow instructions and produce useful assistance. Classically, this involves generating completions from the LLM in response to a query before using a separate reward model to assign a score to the full completion. As an auto-regressive process, the LLM has to take many \"actions\" (selecting individual tokens) and only receives a single, sparse reward at the end of an episode, a setup that is known to be difficult to optimise in traditional reinfor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.00782","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/2402.00782/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":"2026-07-05T07:40:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"c42TCbdjpCwhfYU8P/S1CUjOtugtfXh+jU+U+UdsKmuLGzxAKQx4EPf0N4BuWWC0VZQON8foO0c+60cKAu3FBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T10:53:33.193533Z"},"content_sha256":"6e0f2d3b09e3f1e600fd00458dc8e7b8d303052c17181c5a61cc3f2a225e193d","schema_version":"1.0","event_id":"sha256:6e0f2d3b09e3f1e600fd00458dc8e7b8d303052c17181c5a61cc3f2a225e193d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LGLZBBSUS26O6CMRN7KSPLUTQN/bundle.json","state_url":"https://pith.science/pith/LGLZBBSUS26O6CMRN7KSPLUTQN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LGLZBBSUS26O6CMRN7KSPLUTQN/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":"2026-08-07T10:53:33Z","links":{"resolver":"https://pith.science/pith/LGLZBBSUS26O6CMRN7KSPLUTQN","bundle":"https://pith.science/pith/LGLZBBSUS26O6CMRN7KSPLUTQN/bundle.json","state":"https://pith.science/pith/LGLZBBSUS26O6CMRN7KSPLUTQN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LGLZBBSUS26O6CMRN7KSPLUTQN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:LGLZBBSUS26O6CMRN7KSPLUTQN","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":"f46cf919706d875f4480a757a35e23fa7b95da7e7cfda3053abede483922f4ba","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-01T17:10:35Z","title_canon_sha256":"2163f390390ba469ea48dc253268f9d39b104ad277229e9583fcfd6feff9c179"},"schema_version":"1.0","source":{"id":"2402.00782","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.00782","created_at":"2026-07-05T07:40:17Z"},{"alias_kind":"arxiv_version","alias_value":"2402.00782v1","created_at":"2026-07-05T07:40:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.00782","created_at":"2026-07-05T07:40:17Z"},{"alias_kind":"pith_short_12","alias_value":"LGLZBBSUS26O","created_at":"2026-07-05T07:40:17Z"},{"alias_kind":"pith_short_16","alias_value":"LGLZBBSUS26O6CMR","created_at":"2026-07-05T07:40:17Z"},{"alias_kind":"pith_short_8","alias_value":"LGLZBBSU","created_at":"2026-07-05T07:40:17Z"}],"graph_snapshots":[{"event_id":"sha256:6e0f2d3b09e3f1e600fd00458dc8e7b8d303052c17181c5a61cc3f2a225e193d","target":"graph","created_at":"2026-07-05T07:40:17Z","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/2402.00782/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 been credited as the key advance that has allowed Large Language Models (LLMs) to effectively follow instructions and produce useful assistance. Classically, this involves generating completions from the LLM in response to a query before using a separate reward model to assign a score to the full completion. As an auto-regressive process, the LLM has to take many \"actions\" (selecting individual tokens) and only receives a single, sparse reward at the end of an episode, a setup that is known to be difficult to optimise in traditional reinfor","authors_text":"Alex J. Chan, Hao Sun, Mihaela van der Schaar, Samuel Holt","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-01T17:10:35Z","title":"Dense Reward for Free in Reinforcement Learning from Human Feedback"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.00782","kind":"arxiv","version":1},"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":"sha256:24bb7b1291f696b93574f71365a5c285c06524b7e835f9aeb154ca77a4851f6e","target":"record","created_at":"2026-07-05T07:40:17Z","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":"f46cf919706d875f4480a757a35e23fa7b95da7e7cfda3053abede483922f4ba","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-01T17:10:35Z","title_canon_sha256":"2163f390390ba469ea48dc253268f9d39b104ad277229e9583fcfd6feff9c179"},"schema_version":"1.0","source":{"id":"2402.00782","kind":"arxiv","version":1}},"canonical_sha256":"599790865496bcef09916fd527ae93834a13dd979fa74d0381c661ec02bd9943","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"599790865496bcef09916fd527ae93834a13dd979fa74d0381c661ec02bd9943","first_computed_at":"2026-07-05T07:40:17.617458Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:40:17.617458Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"6ykQwH2NvhXpXBSeQHpAcljSnQXGeSbrHaqnJkF4MU0Atkqp2hwl9Qp+FGi9rgnxvje5XPcN07+Z4OGr5E3CDg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:40:17.617871Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.00782","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:24bb7b1291f696b93574f71365a5c285c06524b7e835f9aeb154ca77a4851f6e","sha256:6e0f2d3b09e3f1e600fd00458dc8e7b8d303052c17181c5a61cc3f2a225e193d"],"state_sha256":"2a746fd5b2170dba7bfeef173767a8ce3590cc12ca15cbfe7a14da5ebc6474a0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HXXh2w3DvMAWyLh4InlYVgY6hkFOa92WqCF5U88fcyYhHsC/PjHglDpnMHtxxBYQJ7WqGmA7AS203PaYjJAqDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T10:53:33.198541Z","bundle_sha256":"6884c930c5c594781d80ccab13e574f5621e585001281ce658c92c457d2268ec"}}