{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:OIH2LAFA3AYNMNPD4GMQCU4Q6I","short_pith_number":"pith:OIH2LAFA","canonical_record":{"source":{"id":"2509.02046","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-02T07:43:22Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"35321aba0ebc26ab3f0249a95845499e857e175ca39e44d5389b5536bac99e7a","abstract_canon_sha256":"1dea35b4bf504722a3df51d3dfc6c4bba8dced032ae4ee44278afb745f870b7b"},"schema_version":"1.0"},"canonical_sha256":"720fa580a0d830d635e3e199015390f214144c560869f50144b4a54521203450","source":{"kind":"arxiv","id":"2509.02046","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.02046","created_at":"2026-07-05T12:05:21Z"},{"alias_kind":"arxiv_version","alias_value":"2509.02046v2","created_at":"2026-07-05T12:05:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.02046","created_at":"2026-07-05T12:05:21Z"},{"alias_kind":"pith_short_12","alias_value":"OIH2LAFA3AYN","created_at":"2026-07-05T12:05:21Z"},{"alias_kind":"pith_short_16","alias_value":"OIH2LAFA3AYNMNPD","created_at":"2026-07-05T12:05:21Z"},{"alias_kind":"pith_short_8","alias_value":"OIH2LAFA","created_at":"2026-07-05T12:05:21Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:OIH2LAFA3AYNMNPD4GMQCU4Q6I","target":"record","payload":{"canonical_record":{"source":{"id":"2509.02046","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-02T07:43:22Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"35321aba0ebc26ab3f0249a95845499e857e175ca39e44d5389b5536bac99e7a","abstract_canon_sha256":"1dea35b4bf504722a3df51d3dfc6c4bba8dced032ae4ee44278afb745f870b7b"},"schema_version":"1.0"},"canonical_sha256":"720fa580a0d830d635e3e199015390f214144c560869f50144b4a54521203450","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:05:21.205428Z","signature_b64":"X+4cbf3XLncQH/Gm1fC0c/XETRVZYkLsOhOyvS8l1TDR7AG2xn4hoKyge+SEY2p69V/196cWKnNnX7Rj25VZBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"720fa580a0d830d635e3e199015390f214144c560869f50144b4a54521203450","last_reissued_at":"2026-07-05T12:05:21.204934Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:05:21.204934Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2509.02046","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":"2026-07-05T12:05:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wxkQU706v3cYmP+rjxHatum4AWoCpisYpsYNsHoB/1Y3TW1N+jNo8GnySbrxyY7ENsT+rCH/6DlKBCpnWLesCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T15:57:19.483811Z"},"content_sha256":"f72c08afc7dff69bfa0d339ab2572bad96268bb97a39157d56c191fff074af19","schema_version":"1.0","event_id":"sha256:f72c08afc7dff69bfa0d339ab2572bad96268bb97a39157d56c191fff074af19"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:OIH2LAFA3AYNMNPD4GMQCU4Q6I","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Fantastic Pretraining Optimizers and Where to Find Them","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"David Hall, Kaiyue Wen, Percy Liang, Tengyu Ma","submitted_at":"2025-09-02T07:43:22Z","abstract_excerpt":"AdamW has long been the dominant optimizer in language model pretraining, despite numerous claims that alternative optimizers offer 1.4 to 2x speedup. We posit that two methodological shortcomings have obscured fair comparisons and hindered practical adoption: (i) unequal hyperparameter tuning and (ii) limited or misleading evaluation setups. To address these two issues, we conduct a systematic study of ten deep learning optimizers across four model scales (0.1B-1.2B parameters) and data-to-model ratios (1-8x the Chinchilla optimum). We find that fair and informative comparisons require rigoro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.02046","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/2509.02046/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":"2026-07-05T12:05:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3yemMZulq0OsYwPlj7n07MI4NFjj395mhj6nMbc3RoGegtn867qtOx9fEm8mhWIWaizFexOsN68uhH9d3+olBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T15:57:19.484314Z"},"content_sha256":"5bffc58539cc7cabae5566843405f7655c900ebc5b73fec490641933d6a8a962","schema_version":"1.0","event_id":"sha256:5bffc58539cc7cabae5566843405f7655c900ebc5b73fec490641933d6a8a962"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OIH2LAFA3AYNMNPD4GMQCU4Q6I/bundle.json","state_url":"https://pith.science/pith/OIH2LAFA3AYNMNPD4GMQCU4Q6I/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OIH2LAFA3AYNMNPD4GMQCU4Q6I/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":"2026-08-15T15:57:19Z","links":{"resolver":"https://pith.science/pith/OIH2LAFA3AYNMNPD4GMQCU4Q6I","bundle":"https://pith.science/pith/OIH2LAFA3AYNMNPD4GMQCU4Q6I/bundle.json","state":"https://pith.science/pith/OIH2LAFA3AYNMNPD4GMQCU4Q6I/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OIH2LAFA3AYNMNPD4GMQCU4Q6I/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:OIH2LAFA3AYNMNPD4GMQCU4Q6I","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":"1dea35b4bf504722a3df51d3dfc6c4bba8dced032ae4ee44278afb745f870b7b","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-02T07:43:22Z","title_canon_sha256":"35321aba0ebc26ab3f0249a95845499e857e175ca39e44d5389b5536bac99e7a"},"schema_version":"1.0","source":{"id":"2509.02046","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.02046","created_at":"2026-07-05T12:05:21Z"},{"alias_kind":"arxiv_version","alias_value":"2509.02046v2","created_at":"2026-07-05T12:05:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.02046","created_at":"2026-07-05T12:05:21Z"},{"alias_kind":"pith_short_12","alias_value":"OIH2LAFA3AYN","created_at":"2026-07-05T12:05:21Z"},{"alias_kind":"pith_short_16","alias_value":"OIH2LAFA3AYNMNPD","created_at":"2026-07-05T12:05:21Z"},{"alias_kind":"pith_short_8","alias_value":"OIH2LAFA","created_at":"2026-07-05T12:05:21Z"}],"graph_snapshots":[{"event_id":"sha256:5bffc58539cc7cabae5566843405f7655c900ebc5b73fec490641933d6a8a962","target":"graph","created_at":"2026-07-05T12:05:21Z","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/2509.02046/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"AdamW has long been the dominant optimizer in language model pretraining, despite numerous claims that alternative optimizers offer 1.4 to 2x speedup. We posit that two methodological shortcomings have obscured fair comparisons and hindered practical adoption: (i) unequal hyperparameter tuning and (ii) limited or misleading evaluation setups. To address these two issues, we conduct a systematic study of ten deep learning optimizers across four model scales (0.1B-1.2B parameters) and data-to-model ratios (1-8x the Chinchilla optimum). We find that fair and informative comparisons require rigoro","authors_text":"David Hall, Kaiyue Wen, Percy Liang, Tengyu Ma","cross_cats":["cs.AI","stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-02T07:43:22Z","title":"Fantastic Pretraining Optimizers and Where to Find Them"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.02046","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:f72c08afc7dff69bfa0d339ab2572bad96268bb97a39157d56c191fff074af19","target":"record","created_at":"2026-07-05T12:05:21Z","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":"1dea35b4bf504722a3df51d3dfc6c4bba8dced032ae4ee44278afb745f870b7b","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-02T07:43:22Z","title_canon_sha256":"35321aba0ebc26ab3f0249a95845499e857e175ca39e44d5389b5536bac99e7a"},"schema_version":"1.0","source":{"id":"2509.02046","kind":"arxiv","version":2}},"canonical_sha256":"720fa580a0d830d635e3e199015390f214144c560869f50144b4a54521203450","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"720fa580a0d830d635e3e199015390f214144c560869f50144b4a54521203450","first_computed_at":"2026-07-05T12:05:21.204934Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:05:21.204934Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"X+4cbf3XLncQH/Gm1fC0c/XETRVZYkLsOhOyvS8l1TDR7AG2xn4hoKyge+SEY2p69V/196cWKnNnX7Rj25VZBw==","signature_status":"signed_v1","signed_at":"2026-07-05T12:05:21.205428Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.02046","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f72c08afc7dff69bfa0d339ab2572bad96268bb97a39157d56c191fff074af19","sha256:5bffc58539cc7cabae5566843405f7655c900ebc5b73fec490641933d6a8a962"],"state_sha256":"7c176ba7cfac44b6e70e8bc9c1fb1e86f5e6fcbfd21519f9a4cf0854eb4edf9f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HVAjsAl+1UBjBbE/Xw88j8TELSTujMVz2rJetHN9NAGnyn6/UXuKU/MjymDCC/0SPbnadYAD4RwFNm75bHzXDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-15T15:57:19.489274Z","bundle_sha256":"8bf82943ebcb27f307651bb9fbf1147578bc948aa5c52536608024fb056e0c64"}}