{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:ONWCAFRN7CFGCRHY6VAJHC6M33","short_pith_number":"pith:ONWCAFRN","canonical_record":{"source":{"id":"2509.01440","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-01T12:50:30Z","cross_cats_sorted":[],"title_canon_sha256":"cd164ff75f2a673340dc50b0bc7ef6816a57e970d3f25db4f87f3b9c22ac2bfe","abstract_canon_sha256":"c8d48fc228084b766b0465ab21ae4916d03abb5c1e5d802a19c6047eacc45ab1"},"schema_version":"1.0"},"canonical_sha256":"736c20162df88a6144f8f540938bccdedb7f6c50988b256860fd29530d647fb2","source":{"kind":"arxiv","id":"2509.01440","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.01440","created_at":"2026-07-05T12:03:00Z"},{"alias_kind":"arxiv_version","alias_value":"2509.01440v1","created_at":"2026-07-05T12:03:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.01440","created_at":"2026-07-05T12:03:00Z"},{"alias_kind":"pith_short_12","alias_value":"ONWCAFRN7CFG","created_at":"2026-07-05T12:03:00Z"},{"alias_kind":"pith_short_16","alias_value":"ONWCAFRN7CFGCRHY","created_at":"2026-07-05T12:03:00Z"},{"alias_kind":"pith_short_8","alias_value":"ONWCAFRN","created_at":"2026-07-05T12:03:00Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:ONWCAFRN7CFGCRHY6VAJHC6M33","target":"record","payload":{"canonical_record":{"source":{"id":"2509.01440","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-01T12:50:30Z","cross_cats_sorted":[],"title_canon_sha256":"cd164ff75f2a673340dc50b0bc7ef6816a57e970d3f25db4f87f3b9c22ac2bfe","abstract_canon_sha256":"c8d48fc228084b766b0465ab21ae4916d03abb5c1e5d802a19c6047eacc45ab1"},"schema_version":"1.0"},"canonical_sha256":"736c20162df88a6144f8f540938bccdedb7f6c50988b256860fd29530d647fb2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:03:00.982825Z","signature_b64":"1D5S9W38HMEgNDs/NwpwqAhqGHVb3FhcEPVfUYGsEV5xLhlyCzxGz41QHjwNMlH/yIz8BQZyMm3ad2EcVL+fAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"736c20162df88a6144f8f540938bccdedb7f6c50988b256860fd29530d647fb2","last_reissued_at":"2026-07-05T12:03:00.982307Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:03:00.982307Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2509.01440","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":"2026-07-05T12:03:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FFugyVHEYSwSK2VrCgsu+QHZjdV4ZAaCF+g4xjXc5KE5776mgxzZ8wBo8Bol0z6KEuIbaHJtq4xW+zNv6y7tDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T02:17:11.530201Z"},"content_sha256":"950c31488fa66e7a334c6bd33f3b5ea8cc58042169c085e2694ef893a6321e58","schema_version":"1.0","event_id":"sha256:950c31488fa66e7a334c6bd33f3b5ea8cc58042169c085e2694ef893a6321e58"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:ONWCAFRN7CFGCRHY6VAJHC6M33","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Benchmarking Optimizers for Large Language Model Pretraining","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Andrei Semenov, Martin Jaggi, Matteo Pagliardini","submitted_at":"2025-09-01T12:50:30Z","abstract_excerpt":"The recent development of Large Language Models (LLMs) has been accompanied by an effervescence of novel ideas and methods to better optimize the loss of deep learning models. Claims from those methods are myriad: from faster convergence to removing reliance on certain hyperparameters. However, the diverse experimental protocols used to validate these claims make direct comparisons between methods challenging. This study presents a comprehensive evaluation of recent optimization techniques across standardized LLM pretraining scenarios, systematically varying model size, batch size, and trainin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.01440","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/2509.01440/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":"2026-07-05T12:03:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tHwtC0YYBNbM3c4c/dMb3jRRKTA/qqi8g37kEplAIjWQHrAQFtd3wIVJVr+hvTw+eGsHEk9Rz0KUw5WAcKzjCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T02:17:11.530895Z"},"content_sha256":"78c10a3be61f8c24c6d7e25b4753fcd29f140561eb6143829119ad002c1c3ae4","schema_version":"1.0","event_id":"sha256:78c10a3be61f8c24c6d7e25b4753fcd29f140561eb6143829119ad002c1c3ae4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ONWCAFRN7CFGCRHY6VAJHC6M33/bundle.json","state_url":"https://pith.science/pith/ONWCAFRN7CFGCRHY6VAJHC6M33/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ONWCAFRN7CFGCRHY6VAJHC6M33/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":"2026-08-11T02:17:11Z","links":{"resolver":"https://pith.science/pith/ONWCAFRN7CFGCRHY6VAJHC6M33","bundle":"https://pith.science/pith/ONWCAFRN7CFGCRHY6VAJHC6M33/bundle.json","state":"https://pith.science/pith/ONWCAFRN7CFGCRHY6VAJHC6M33/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ONWCAFRN7CFGCRHY6VAJHC6M33/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ONWCAFRN7CFGCRHY6VAJHC6M33","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":"c8d48fc228084b766b0465ab21ae4916d03abb5c1e5d802a19c6047eacc45ab1","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-01T12:50:30Z","title_canon_sha256":"cd164ff75f2a673340dc50b0bc7ef6816a57e970d3f25db4f87f3b9c22ac2bfe"},"schema_version":"1.0","source":{"id":"2509.01440","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.01440","created_at":"2026-07-05T12:03:00Z"},{"alias_kind":"arxiv_version","alias_value":"2509.01440v1","created_at":"2026-07-05T12:03:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.01440","created_at":"2026-07-05T12:03:00Z"},{"alias_kind":"pith_short_12","alias_value":"ONWCAFRN7CFG","created_at":"2026-07-05T12:03:00Z"},{"alias_kind":"pith_short_16","alias_value":"ONWCAFRN7CFGCRHY","created_at":"2026-07-05T12:03:00Z"},{"alias_kind":"pith_short_8","alias_value":"ONWCAFRN","created_at":"2026-07-05T12:03:00Z"}],"graph_snapshots":[{"event_id":"sha256:78c10a3be61f8c24c6d7e25b4753fcd29f140561eb6143829119ad002c1c3ae4","target":"graph","created_at":"2026-07-05T12:03:00Z","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/2509.01440/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The recent development of Large Language Models (LLMs) has been accompanied by an effervescence of novel ideas and methods to better optimize the loss of deep learning models. Claims from those methods are myriad: from faster convergence to removing reliance on certain hyperparameters. However, the diverse experimental protocols used to validate these claims make direct comparisons between methods challenging. This study presents a comprehensive evaluation of recent optimization techniques across standardized LLM pretraining scenarios, systematically varying model size, batch size, and trainin","authors_text":"Andrei Semenov, Martin Jaggi, Matteo Pagliardini","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-01T12:50:30Z","title":"Benchmarking Optimizers for Large Language Model Pretraining"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.01440","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":"sha256:950c31488fa66e7a334c6bd33f3b5ea8cc58042169c085e2694ef893a6321e58","target":"record","created_at":"2026-07-05T12:03:00Z","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":"c8d48fc228084b766b0465ab21ae4916d03abb5c1e5d802a19c6047eacc45ab1","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-01T12:50:30Z","title_canon_sha256":"cd164ff75f2a673340dc50b0bc7ef6816a57e970d3f25db4f87f3b9c22ac2bfe"},"schema_version":"1.0","source":{"id":"2509.01440","kind":"arxiv","version":1}},"canonical_sha256":"736c20162df88a6144f8f540938bccdedb7f6c50988b256860fd29530d647fb2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"736c20162df88a6144f8f540938bccdedb7f6c50988b256860fd29530d647fb2","first_computed_at":"2026-07-05T12:03:00.982307Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:03:00.982307Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1D5S9W38HMEgNDs/NwpwqAhqGHVb3FhcEPVfUYGsEV5xLhlyCzxGz41QHjwNMlH/yIz8BQZyMm3ad2EcVL+fAg==","signature_status":"signed_v1","signed_at":"2026-07-05T12:03:00.982825Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.01440","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:950c31488fa66e7a334c6bd33f3b5ea8cc58042169c085e2694ef893a6321e58","sha256:78c10a3be61f8c24c6d7e25b4753fcd29f140561eb6143829119ad002c1c3ae4"],"state_sha256":"7b975edb45d7bd1f5b82a2200b6755f2f44d4b1cd90012fb40f63dd5e94f2c19"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KzqBnUkW1Yiva6vrcnj6rAoDSqiG9hq1alCzHHclCh1okfD0S71TZIuQ7xUFcLboMc/ygLziEsNH0IMa9Z+mCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T02:17:11.538449Z","bundle_sha256":"973b8a2d906c1a5250ea74c9d6835b250f64f6c55abce25c94688cb760fc8141"}}