{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:RU63PLFCHYAMZAFGLLDAOQA5PV","short_pith_number":"pith:RU63PLFC","canonical_record":{"source":{"id":"2406.07168","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-11T11:20:05Z","cross_cats_sorted":[],"title_canon_sha256":"fdb85e2baab4a4713a8378c72438b92aacc673dd553e33aed4aa605dfa443896","abstract_canon_sha256":"f177b43db9b2331daa5042733f3b2e61d99207f035a814c3f02cfdb4fd41f712"},"schema_version":"1.0"},"canonical_sha256":"8d3db7aca23e00cc80a65ac607401d7d5bab7a453cd324d8800c97781ba3facd","source":{"kind":"arxiv","id":"2406.07168","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.07168","created_at":"2026-07-05T08:30:19Z"},{"alias_kind":"arxiv_version","alias_value":"2406.07168v1","created_at":"2026-07-05T08:30:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.07168","created_at":"2026-07-05T08:30:19Z"},{"alias_kind":"pith_short_12","alias_value":"RU63PLFCHYAM","created_at":"2026-07-05T08:30:19Z"},{"alias_kind":"pith_short_16","alias_value":"RU63PLFCHYAMZAFG","created_at":"2026-07-05T08:30:19Z"},{"alias_kind":"pith_short_8","alias_value":"RU63PLFC","created_at":"2026-07-05T08:30:19Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:RU63PLFCHYAMZAFGLLDAOQA5PV","target":"record","payload":{"canonical_record":{"source":{"id":"2406.07168","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-11T11:20:05Z","cross_cats_sorted":[],"title_canon_sha256":"fdb85e2baab4a4713a8378c72438b92aacc673dd553e33aed4aa605dfa443896","abstract_canon_sha256":"f177b43db9b2331daa5042733f3b2e61d99207f035a814c3f02cfdb4fd41f712"},"schema_version":"1.0"},"canonical_sha256":"8d3db7aca23e00cc80a65ac607401d7d5bab7a453cd324d8800c97781ba3facd","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:30:19.421260Z","signature_b64":"mgVD6YsRE1yA2pms4BuXTp3/ZYjNL+PQ2KmasZBH1XsnSZkzyFq/vxGjlrmVza6Iy0WcUUk/C67lMc5Mo5osDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8d3db7aca23e00cc80a65ac607401d7d5bab7a453cd324d8800c97781ba3facd","last_reissued_at":"2026-07-05T08:30:19.420739Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:30:19.420739Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.07168","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":"2026-07-05T08:30:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+J7JHQ6DHJL9EkpqwxOx4VxfWil36xW48gvwWA8JrijS+jQWuzNgH4DbUAn+GF1BB87Ul2OnR+zx0jGrn7vjDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T09:27:57.918343Z"},"content_sha256":"cf343d3af970ccc0fbf5b3d2e77d9b09d86fabc3695f134f8c21ca8a6242f2db","schema_version":"1.0","event_id":"sha256:cf343d3af970ccc0fbf5b3d2e77d9b09d86fabc3695f134f8c21ca8a6242f2db"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:RU63PLFCHYAMZAFGLLDAOQA5PV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Teaching Language Models to Self-Improve by Learning from Language Feedback","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chi Hu, Hang Cao, Jingbo Zhu, Tong Xiao, Yimin Hu","submitted_at":"2024-06-11T11:20:05Z","abstract_excerpt":"Aligning Large Language Models (LLMs) with human intentions and values is crucial yet challenging. Current methods primarily rely on human preferences, which are costly and insufficient in capturing nuanced feedback expressed in natural language. In this paper, we present Self-Refinement Tuning (SRT), a method that leverages model feedback for alignment, thereby reducing reliance on human annotations. SRT uses a base language model (e.g., Tulu2) to generate initial responses, which are critiqued and refined by a more advanced model (e.g., GPT-4-Turbo). This process enables the base model to se"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.07168","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/2406.07168/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":"2026-07-05T08:30:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ylfaSEoj+mjTgE1HxkyKAUF12rqxWY60iw6MiN2U45dnU/FbbjFrZQxlUc5AVYygbasRYBICznmKJhcfiZB8Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T09:27:57.918982Z"},"content_sha256":"ae60dbad7b8850124ff0da32984de8d915247117a936c62f052adb538b2618d9","schema_version":"1.0","event_id":"sha256:ae60dbad7b8850124ff0da32984de8d915247117a936c62f052adb538b2618d9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RU63PLFCHYAMZAFGLLDAOQA5PV/bundle.json","state_url":"https://pith.science/pith/RU63PLFCHYAMZAFGLLDAOQA5PV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RU63PLFCHYAMZAFGLLDAOQA5PV/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":"2026-08-12T09:27:57Z","links":{"resolver":"https://pith.science/pith/RU63PLFCHYAMZAFGLLDAOQA5PV","bundle":"https://pith.science/pith/RU63PLFCHYAMZAFGLLDAOQA5PV/bundle.json","state":"https://pith.science/pith/RU63PLFCHYAMZAFGLLDAOQA5PV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RU63PLFCHYAMZAFGLLDAOQA5PV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:RU63PLFCHYAMZAFGLLDAOQA5PV","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":"f177b43db9b2331daa5042733f3b2e61d99207f035a814c3f02cfdb4fd41f712","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-11T11:20:05Z","title_canon_sha256":"fdb85e2baab4a4713a8378c72438b92aacc673dd553e33aed4aa605dfa443896"},"schema_version":"1.0","source":{"id":"2406.07168","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.07168","created_at":"2026-07-05T08:30:19Z"},{"alias_kind":"arxiv_version","alias_value":"2406.07168v1","created_at":"2026-07-05T08:30:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.07168","created_at":"2026-07-05T08:30:19Z"},{"alias_kind":"pith_short_12","alias_value":"RU63PLFCHYAM","created_at":"2026-07-05T08:30:19Z"},{"alias_kind":"pith_short_16","alias_value":"RU63PLFCHYAMZAFG","created_at":"2026-07-05T08:30:19Z"},{"alias_kind":"pith_short_8","alias_value":"RU63PLFC","created_at":"2026-07-05T08:30:19Z"}],"graph_snapshots":[{"event_id":"sha256:ae60dbad7b8850124ff0da32984de8d915247117a936c62f052adb538b2618d9","target":"graph","created_at":"2026-07-05T08:30:19Z","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/2406.07168/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Aligning Large Language Models (LLMs) with human intentions and values is crucial yet challenging. Current methods primarily rely on human preferences, which are costly and insufficient in capturing nuanced feedback expressed in natural language. In this paper, we present Self-Refinement Tuning (SRT), a method that leverages model feedback for alignment, thereby reducing reliance on human annotations. SRT uses a base language model (e.g., Tulu2) to generate initial responses, which are critiqued and refined by a more advanced model (e.g., GPT-4-Turbo). This process enables the base model to se","authors_text":"Chi Hu, Hang Cao, Jingbo Zhu, Tong Xiao, Yimin Hu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-11T11:20:05Z","title":"Teaching Language Models to Self-Improve by Learning from Language Feedback"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.07168","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:cf343d3af970ccc0fbf5b3d2e77d9b09d86fabc3695f134f8c21ca8a6242f2db","target":"record","created_at":"2026-07-05T08:30:19Z","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":"f177b43db9b2331daa5042733f3b2e61d99207f035a814c3f02cfdb4fd41f712","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-11T11:20:05Z","title_canon_sha256":"fdb85e2baab4a4713a8378c72438b92aacc673dd553e33aed4aa605dfa443896"},"schema_version":"1.0","source":{"id":"2406.07168","kind":"arxiv","version":1}},"canonical_sha256":"8d3db7aca23e00cc80a65ac607401d7d5bab7a453cd324d8800c97781ba3facd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8d3db7aca23e00cc80a65ac607401d7d5bab7a453cd324d8800c97781ba3facd","first_computed_at":"2026-07-05T08:30:19.420739Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:30:19.420739Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"mgVD6YsRE1yA2pms4BuXTp3/ZYjNL+PQ2KmasZBH1XsnSZkzyFq/vxGjlrmVza6Iy0WcUUk/C67lMc5Mo5osDA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:30:19.421260Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.07168","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:cf343d3af970ccc0fbf5b3d2e77d9b09d86fabc3695f134f8c21ca8a6242f2db","sha256:ae60dbad7b8850124ff0da32984de8d915247117a936c62f052adb538b2618d9"],"state_sha256":"ea27229257e3761b2f4020b0ea7c28e6336ff274e0c322c5c81cbfba98c2e9c9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BE/qDfwn+eb8Ed5TXYvQ6QZBcEflCFqMIvrObviAq2IGUs8ttHiYeQ9sCUwj0VgALjnknN4fhu8p86/4N2vcCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-12T09:27:57.924081Z","bundle_sha256":"404b231808b6972c165e99c10ea9e2d4f4256823484ba6b784c5aa1c468a4098"}}