{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:S4YJM6DB5GAKJSEQM44DDLNQSF","short_pith_number":"pith:S4YJM6DB","canonical_record":{"source":{"id":"2304.05302","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-04-11T15:53:40Z","cross_cats_sorted":[],"title_canon_sha256":"70e2728e70418462d0ff2ebdbc07695529b1af739993c727a3bb89d857ce7d1f","abstract_canon_sha256":"47fbedcf1345809619c1f20df47e4298d460ca1f4df403ad4189e03427e4fc90"},"schema_version":"1.0"},"canonical_sha256":"9730967861e980a4c890673831adb09178b63c3c91681b1f06cfd98de96d416c","source":{"kind":"arxiv","id":"2304.05302","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2304.05302","created_at":"2026-07-05T06:58:08Z"},{"alias_kind":"arxiv_version","alias_value":"2304.05302v3","created_at":"2026-07-05T06:58:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2304.05302","created_at":"2026-07-05T06:58:08Z"},{"alias_kind":"pith_short_12","alias_value":"S4YJM6DB5GAK","created_at":"2026-07-05T06:58:08Z"},{"alias_kind":"pith_short_16","alias_value":"S4YJM6DB5GAKJSEQ","created_at":"2026-07-05T06:58:08Z"},{"alias_kind":"pith_short_8","alias_value":"S4YJM6DB","created_at":"2026-07-05T06:58:08Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:S4YJM6DB5GAKJSEQM44DDLNQSF","target":"record","payload":{"canonical_record":{"source":{"id":"2304.05302","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-04-11T15:53:40Z","cross_cats_sorted":[],"title_canon_sha256":"70e2728e70418462d0ff2ebdbc07695529b1af739993c727a3bb89d857ce7d1f","abstract_canon_sha256":"47fbedcf1345809619c1f20df47e4298d460ca1f4df403ad4189e03427e4fc90"},"schema_version":"1.0"},"canonical_sha256":"9730967861e980a4c890673831adb09178b63c3c91681b1f06cfd98de96d416c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:58:08.990935Z","signature_b64":"0yDXfpw9mqFcD0OJPP8Nd5zRDJ3OlG7yLvy3H4Tu+0k+h/eckjrhE2tH1nSKLmipa5hWCVa/xg5oc5lKCjITDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9730967861e980a4c890673831adb09178b63c3c91681b1f06cfd98de96d416c","last_reissued_at":"2026-07-05T06:58:08.990437Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:58:08.990437Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2304.05302","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":"2026-07-05T06:58:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qshsFaLRPlwpbgn9FJFtzExFWmqZ+oV8nomH5GuHH23vKl+Iq88UKAKK+EYiyDtg3phK7i4Cv7aMfz0+p6SYDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T22:37:49.056998Z"},"content_sha256":"74159f75be4968b0ef288805dd482d15607471314d1e699ad76217cd8224db30","schema_version":"1.0","event_id":"sha256:74159f75be4968b0ef288805dd482d15607471314d1e699ad76217cd8224db30"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:S4YJM6DB5GAKJSEQM44DDLNQSF","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"RRHF: Rank Responses to Align Language Models with Human Feedback without tears","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chuanqi Tan, Fei Huang, Hongyi Yuan, Songfang Huang, Wei Wang, Zheng Yuan","submitted_at":"2023-04-11T15:53:40Z","abstract_excerpt":"Reinforcement Learning from Human Feedback (RLHF) facilitates the alignment of large language models with human preferences, significantly enhancing the quality of interactions between humans and models. InstructGPT implements RLHF through several stages, including Supervised Fine-Tuning (SFT), reward model training, and Proximal Policy Optimization (PPO). However, PPO is sensitive to hyperparameters and requires multiple models in its standard implementation, making it hard to train and scale up to larger parameter counts. In contrast, we propose a novel learning paradigm called RRHF, which s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2304.05302","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/2304.05302/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":"2026-07-05T06:58:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qNVr8e38msBBzD3udjFN3fPR1HvX3ckEUG83RlB1nqXzu3NbJbbTnf4YO+WXv/PK+zwnogPfJ7X5zxTM6j6dCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T22:37:49.057530Z"},"content_sha256":"82b9f574c1c55255972e8a39b6a4a51706ffe68419c7f94fd16f10f2321cca23","schema_version":"1.0","event_id":"sha256:82b9f574c1c55255972e8a39b6a4a51706ffe68419c7f94fd16f10f2321cca23"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/S4YJM6DB5GAKJSEQM44DDLNQSF/bundle.json","state_url":"https://pith.science/pith/S4YJM6DB5GAKJSEQM44DDLNQSF/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/S4YJM6DB5GAKJSEQM44DDLNQSF/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":"2026-08-05T22:37:49Z","links":{"resolver":"https://pith.science/pith/S4YJM6DB5GAKJSEQM44DDLNQSF","bundle":"https://pith.science/pith/S4YJM6DB5GAKJSEQM44DDLNQSF/bundle.json","state":"https://pith.science/pith/S4YJM6DB5GAKJSEQM44DDLNQSF/state.json","well_known_bundle":"https://pith.science/.well-known/pith/S4YJM6DB5GAKJSEQM44DDLNQSF/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:S4YJM6DB5GAKJSEQM44DDLNQSF","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":"47fbedcf1345809619c1f20df47e4298d460ca1f4df403ad4189e03427e4fc90","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-04-11T15:53:40Z","title_canon_sha256":"70e2728e70418462d0ff2ebdbc07695529b1af739993c727a3bb89d857ce7d1f"},"schema_version":"1.0","source":{"id":"2304.05302","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2304.05302","created_at":"2026-07-05T06:58:08Z"},{"alias_kind":"arxiv_version","alias_value":"2304.05302v3","created_at":"2026-07-05T06:58:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2304.05302","created_at":"2026-07-05T06:58:08Z"},{"alias_kind":"pith_short_12","alias_value":"S4YJM6DB5GAK","created_at":"2026-07-05T06:58:08Z"},{"alias_kind":"pith_short_16","alias_value":"S4YJM6DB5GAKJSEQ","created_at":"2026-07-05T06:58:08Z"},{"alias_kind":"pith_short_8","alias_value":"S4YJM6DB","created_at":"2026-07-05T06:58:08Z"}],"graph_snapshots":[{"event_id":"sha256:82b9f574c1c55255972e8a39b6a4a51706ffe68419c7f94fd16f10f2321cca23","target":"graph","created_at":"2026-07-05T06:58:08Z","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/2304.05302/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement Learning from Human Feedback (RLHF) facilitates the alignment of large language models with human preferences, significantly enhancing the quality of interactions between humans and models. InstructGPT implements RLHF through several stages, including Supervised Fine-Tuning (SFT), reward model training, and Proximal Policy Optimization (PPO). However, PPO is sensitive to hyperparameters and requires multiple models in its standard implementation, making it hard to train and scale up to larger parameter counts. In contrast, we propose a novel learning paradigm called RRHF, which s","authors_text":"Chuanqi Tan, Fei Huang, Hongyi Yuan, Songfang Huang, Wei Wang, Zheng Yuan","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-04-11T15:53:40Z","title":"RRHF: Rank Responses to Align Language Models with Human Feedback without tears"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2304.05302","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":"sha256:74159f75be4968b0ef288805dd482d15607471314d1e699ad76217cd8224db30","target":"record","created_at":"2026-07-05T06:58:08Z","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":"47fbedcf1345809619c1f20df47e4298d460ca1f4df403ad4189e03427e4fc90","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-04-11T15:53:40Z","title_canon_sha256":"70e2728e70418462d0ff2ebdbc07695529b1af739993c727a3bb89d857ce7d1f"},"schema_version":"1.0","source":{"id":"2304.05302","kind":"arxiv","version":3}},"canonical_sha256":"9730967861e980a4c890673831adb09178b63c3c91681b1f06cfd98de96d416c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9730967861e980a4c890673831adb09178b63c3c91681b1f06cfd98de96d416c","first_computed_at":"2026-07-05T06:58:08.990437Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:58:08.990437Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"0yDXfpw9mqFcD0OJPP8Nd5zRDJ3OlG7yLvy3H4Tu+0k+h/eckjrhE2tH1nSKLmipa5hWCVa/xg5oc5lKCjITDw==","signature_status":"signed_v1","signed_at":"2026-07-05T06:58:08.990935Z","signed_message":"canonical_sha256_bytes"},"source_id":"2304.05302","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:74159f75be4968b0ef288805dd482d15607471314d1e699ad76217cd8224db30","sha256:82b9f574c1c55255972e8a39b6a4a51706ffe68419c7f94fd16f10f2321cca23"],"state_sha256":"38a3f1cfd59c0f54b707bf100ddf1c7ca525acbc7117a96184d7649715dfb465"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"S3z8MMVkM7DVnIR9M90jtBfP++UY7WYnzm9kSOSlx2FCEcp1QPM++qCSl5LUwMlbPBf7HX3kshPF0PokfghzAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T22:37:49.061664Z","bundle_sha256":"d980f42dc3c31cbd1edcc296f927bcb73e3f2aa682f6707aabd9de96733aa10a"}}