{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:ASFYTQH3XFGPOZFXEVB3MB7U6L","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":"8d55dcf5f3772afaee6ff92cccb34842f072e268efc47f0aa90d2b6aac6c55ac","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-02-27T17:11:15Z","title_canon_sha256":"29590c531c32a771af9098899c1b11716a5ab715df1c011c63f23145d0657a7d"},"schema_version":"1.0","source":{"id":"2302.13971","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2302.13971","created_at":"2026-07-05T05:45:52Z"},{"alias_kind":"arxiv_version","alias_value":"2302.13971v1","created_at":"2026-07-05T05:45:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.13971","created_at":"2026-07-05T05:45:52Z"},{"alias_kind":"pith_short_12","alias_value":"ASFYTQH3XFGP","created_at":"2026-07-05T05:45:52Z"},{"alias_kind":"pith_short_16","alias_value":"ASFYTQH3XFGPOZFX","created_at":"2026-07-05T05:45:52Z"},{"alias_kind":"pith_short_8","alias_value":"ASFYTQH3","created_at":"2026-07-05T05:45:52Z"}],"graph_snapshots":[{"event_id":"sha256:2e62695c8f79526a66d887ecfd17c68587e2facbdfc57160dc43c5903732ab69","target":"graph","created_at":"2026-07-05T05:45:52Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"LLaMA-13B outperforms GPT-3 (175B) on most benchmarks, and LLaMA-65B is competitive with the best models, Chinchilla-70B and PaLM-540B."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The evaluation benchmarks used are fair and representative measures of capability that do not favor models trained on the specific public data mixtures chosen by the authors."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"LLaMA models (7B-65B params) trained only on public data outperform GPT-3 (175B) on most benchmarks and match top closed models like Chinchilla-70B and PaLM-540B."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"LLaMA models trained only on public data outperform GPT-3 despite fewer parameters."}],"snapshot_sha256":"377522040569474d7a7348b3a4b5f6dca531347958ebd39ea2fe66fec3c6b1ee"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2302.13971/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce LLaMA, a collection of foundation language models ranging from 7B to 65B parameters. We train our models on trillions of tokens, and show that it is possible to train state-of-the-art models using publicly available datasets exclusively, without resorting to proprietary and inaccessible datasets. In particular, LLaMA-13B outperforms GPT-3 (175B) on most benchmarks, and LLaMA-65B is competitive with the best models, Chinchilla-70B and PaLM-540B. We release all our models to the research community.","authors_text":"Armand Joulin, Aurelien Rodriguez, Baptiste Rozi\\`ere, Edouard Grave, Eric Hambro, Faisal Azhar, Gautier Izacard, Guillaume Lample, Hugo Touvron, Marie-Anne Lachaux, Naman Goyal, Thibaut Lavril, Timoth\\'ee Lacroix, Xavier Martinet","cross_cats":[],"headline":"LLaMA models trained only on public data outperform GPT-3 despite fewer parameters.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-02-27T17:11:15Z","title":"LLaMA: Open and Efficient Foundation Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.13971","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-24T09:34:01.387284Z","id":"3d88d2ac-6e3b-4d62-ae36-84c90c9a04ad","model_set":{"reader":"grok-4.3"},"one_line_summary":"LLaMA models (7B-65B params) trained only on public data outperform GPT-3 (175B) on most benchmarks and match top closed models like Chinchilla-70B and PaLM-540B.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"LLaMA models trained only on public data outperform GPT-3 despite fewer parameters.","strongest_claim":"LLaMA-13B outperforms GPT-3 (175B) on most benchmarks, and LLaMA-65B is competitive with the best models, Chinchilla-70B and PaLM-540B.","weakest_assumption":"The evaluation benchmarks used are fair and representative measures of capability that do not favor models trained on the specific public data mixtures chosen by the authors."}},"verdict_id":"3d88d2ac-6e3b-4d62-ae36-84c90c9a04ad"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:b4c4ae3a3b29558c16a1ea8de051a83ebf6ad639c6738b2ae8eec67f856b5d40","target":"record","created_at":"2026-07-05T05:45:52Z","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":"8d55dcf5f3772afaee6ff92cccb34842f072e268efc47f0aa90d2b6aac6c55ac","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-02-27T17:11:15Z","title_canon_sha256":"29590c531c32a771af9098899c1b11716a5ab715df1c011c63f23145d0657a7d"},"schema_version":"1.0","source":{"id":"2302.13971","kind":"arxiv","version":1}},"canonical_sha256":"048b89c0fbb94cf764b72543b607f4f2d8d4899204bcf59329c1a3fecb331deb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"048b89c0fbb94cf764b72543b607f4f2d8d4899204bcf59329c1a3fecb331deb","first_computed_at":"2026-07-05T05:45:52.996357Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:45:52.996357Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"pwFLcIlIZWPcCd1dqMdx1UojeZ/KLvudfVO3191njYuM/cVtQsgqJsnHwxydFimyWoc4ZsbckeoWM7G9YimZBg==","signature_status":"signed_v1","signed_at":"2026-07-05T05:45:52.996879Z","signed_message":"canonical_sha256_bytes"},"source_id":"2302.13971","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b4c4ae3a3b29558c16a1ea8de051a83ebf6ad639c6738b2ae8eec67f856b5d40","sha256:2e62695c8f79526a66d887ecfd17c68587e2facbdfc57160dc43c5903732ab69"],"state_sha256":"eef7b697ca8987669d1f40898bbc9eb9c23ba14991285d73cf1f51fe8fdbf348"}