{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:ZO75ETUXPBJT2WQXHW7E6OLSQ5","short_pith_number":"pith:ZO75ETUX","canonical_record":{"source":{"id":"2403.10131","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-15T09:26:02Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ef7e4a056c38bc56bc8292d4c3467a9d127085b9c9178330023d07de0513f6c0","abstract_canon_sha256":"87a7a07bf34a1a0da3ee321ff78d981fb05657af8b78822f214507b0f34990b0"},"schema_version":"1.0"},"canonical_sha256":"cbbfd24e9778533d5a173dbe4f3972874eeb4a32cf8b61f3828add93c1ef24f2","source":{"kind":"arxiv","id":"2403.10131","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.10131","created_at":"2026-07-05T08:27:47Z"},{"alias_kind":"arxiv_version","alias_value":"2403.10131v2","created_at":"2026-07-05T08:27:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.10131","created_at":"2026-07-05T08:27:47Z"},{"alias_kind":"pith_short_12","alias_value":"ZO75ETUXPBJT","created_at":"2026-07-05T08:27:47Z"},{"alias_kind":"pith_short_16","alias_value":"ZO75ETUXPBJT2WQX","created_at":"2026-07-05T08:27:47Z"},{"alias_kind":"pith_short_8","alias_value":"ZO75ETUX","created_at":"2026-07-05T08:27:47Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:ZO75ETUXPBJT2WQXHW7E6OLSQ5","target":"record","payload":{"canonical_record":{"source":{"id":"2403.10131","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-15T09:26:02Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ef7e4a056c38bc56bc8292d4c3467a9d127085b9c9178330023d07de0513f6c0","abstract_canon_sha256":"87a7a07bf34a1a0da3ee321ff78d981fb05657af8b78822f214507b0f34990b0"},"schema_version":"1.0"},"canonical_sha256":"cbbfd24e9778533d5a173dbe4f3972874eeb4a32cf8b61f3828add93c1ef24f2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:27:47.462140Z","signature_b64":"xvrlVqMFb/TOgkbPaF5NrlMjBVyeOb1F6fOHvMUI+V3uoMOTW3IJuH6cgLwdcWaiYYLboYvqIdRC/6nb9cimAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cbbfd24e9778533d5a173dbe4f3972874eeb4a32cf8b61f3828add93c1ef24f2","last_reissued_at":"2026-07-05T08:27:47.461594Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:27:47.461594Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2403.10131","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":"2026-07-05T08:27:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9gEUy+HvaS+wEoYf7H06zH0YmHGjUUr37rF74KwigSGwdrQx+hZNJFf3VIaZbZ9NajumOG0ZT7Pvrf/Ysvc0Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T00:55:07.057876Z"},"content_sha256":"e7e1555dc1225d60981168f255266d23387d52b585de8f123121cc9b905f0a8a","schema_version":"1.0","event_id":"sha256:e7e1555dc1225d60981168f255266d23387d52b585de8f123121cc9b905f0a8a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:ZO75ETUXPBJT2WQXHW7E6OLSQ5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"RAFT: Adapting Language Model to Domain Specific RAG","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ion Stoica, Joseph E. Gonzalez, Matei Zaharia, Naman Jain, Sheng Shen, Shishir G. Patil, Tianjun Zhang","submitted_at":"2024-03-15T09:26:02Z","abstract_excerpt":"Pretraining Large Language Models (LLMs) on large corpora of textual data is now a standard paradigm. When using these LLMs for many downstream applications, it is common to additionally bake in new knowledge (e.g., time-critical news, or private domain knowledge) into the pretrained model either through RAG-based-prompting, or fine-tuning. However, the optimal methodology for the model to gain such new knowledge remains an open question. In this paper, we present Retrieval Augmented FineTuning (RAFT), a training recipe that improves the model's ability to answer questions in a \"open-book\" in-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.10131","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/2403.10131/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":"2026-07-05T08:27:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1FYpiHuvudm/fSgcdK0DXQbOQ1ncRGMFBNo4aUdKUfR0eR2xerNq4lnAD1Zm46J3XVr5VZCmslKikneQYVmiAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T00:55:07.058204Z"},"content_sha256":"b2970ef7c1a3bda3bf1f32d7e630928d934191e7b6187e69347b06b5a8946809","schema_version":"1.0","event_id":"sha256:b2970ef7c1a3bda3bf1f32d7e630928d934191e7b6187e69347b06b5a8946809"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZO75ETUXPBJT2WQXHW7E6OLSQ5/bundle.json","state_url":"https://pith.science/pith/ZO75ETUXPBJT2WQXHW7E6OLSQ5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZO75ETUXPBJT2WQXHW7E6OLSQ5/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":"2026-08-19T00:55:07Z","links":{"resolver":"https://pith.science/pith/ZO75ETUXPBJT2WQXHW7E6OLSQ5","bundle":"https://pith.science/pith/ZO75ETUXPBJT2WQXHW7E6OLSQ5/bundle.json","state":"https://pith.science/pith/ZO75ETUXPBJT2WQXHW7E6OLSQ5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZO75ETUXPBJT2WQXHW7E6OLSQ5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:ZO75ETUXPBJT2WQXHW7E6OLSQ5","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":"87a7a07bf34a1a0da3ee321ff78d981fb05657af8b78822f214507b0f34990b0","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-15T09:26:02Z","title_canon_sha256":"ef7e4a056c38bc56bc8292d4c3467a9d127085b9c9178330023d07de0513f6c0"},"schema_version":"1.0","source":{"id":"2403.10131","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.10131","created_at":"2026-07-05T08:27:47Z"},{"alias_kind":"arxiv_version","alias_value":"2403.10131v2","created_at":"2026-07-05T08:27:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.10131","created_at":"2026-07-05T08:27:47Z"},{"alias_kind":"pith_short_12","alias_value":"ZO75ETUXPBJT","created_at":"2026-07-05T08:27:47Z"},{"alias_kind":"pith_short_16","alias_value":"ZO75ETUXPBJT2WQX","created_at":"2026-07-05T08:27:47Z"},{"alias_kind":"pith_short_8","alias_value":"ZO75ETUX","created_at":"2026-07-05T08:27:47Z"}],"graph_snapshots":[{"event_id":"sha256:b2970ef7c1a3bda3bf1f32d7e630928d934191e7b6187e69347b06b5a8946809","target":"graph","created_at":"2026-07-05T08:27:47Z","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/2403.10131/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Pretraining Large Language Models (LLMs) on large corpora of textual data is now a standard paradigm. When using these LLMs for many downstream applications, it is common to additionally bake in new knowledge (e.g., time-critical news, or private domain knowledge) into the pretrained model either through RAG-based-prompting, or fine-tuning. However, the optimal methodology for the model to gain such new knowledge remains an open question. In this paper, we present Retrieval Augmented FineTuning (RAFT), a training recipe that improves the model's ability to answer questions in a \"open-book\" in-","authors_text":"Ion Stoica, Joseph E. Gonzalez, Matei Zaharia, Naman Jain, Sheng Shen, Shishir G. Patil, Tianjun Zhang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-15T09:26:02Z","title":"RAFT: Adapting Language Model to Domain Specific RAG"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.10131","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:e7e1555dc1225d60981168f255266d23387d52b585de8f123121cc9b905f0a8a","target":"record","created_at":"2026-07-05T08:27:47Z","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":"87a7a07bf34a1a0da3ee321ff78d981fb05657af8b78822f214507b0f34990b0","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-15T09:26:02Z","title_canon_sha256":"ef7e4a056c38bc56bc8292d4c3467a9d127085b9c9178330023d07de0513f6c0"},"schema_version":"1.0","source":{"id":"2403.10131","kind":"arxiv","version":2}},"canonical_sha256":"cbbfd24e9778533d5a173dbe4f3972874eeb4a32cf8b61f3828add93c1ef24f2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cbbfd24e9778533d5a173dbe4f3972874eeb4a32cf8b61f3828add93c1ef24f2","first_computed_at":"2026-07-05T08:27:47.461594Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:27:47.461594Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"xvrlVqMFb/TOgkbPaF5NrlMjBVyeOb1F6fOHvMUI+V3uoMOTW3IJuH6cgLwdcWaiYYLboYvqIdRC/6nb9cimAw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:27:47.462140Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.10131","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e7e1555dc1225d60981168f255266d23387d52b585de8f123121cc9b905f0a8a","sha256:b2970ef7c1a3bda3bf1f32d7e630928d934191e7b6187e69347b06b5a8946809"],"state_sha256":"1aef63fbb8d196c495f037a8dc7baf12f17f401a4ff7b6efe51de800ca817602"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xS54PyOBoRhfeTBSJWDmRa1A1TcyVLYQxtnnBk9laCXjUuOjKQ0z+Fep9uBe5mqe9HozS6qIuFCWSG5CLW5nBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-19T00:55:07.061622Z","bundle_sha256":"deedc5b317ebb6f55d76665836769fce3765669a056196e8eda8aa43b56017d6"}}