{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:6CIKL56FVSUUEXGNXX4IYWI7JP","short_pith_number":"pith:6CIKL56F","schema_version":"1.0","canonical_sha256":"f090a5f7c5aca9425ccdbdf88c591f4bd58232011e39077204f588f4ec444c29","source":{"kind":"arxiv","id":"2306.12517","version":1},"attestation_state":"computed","paper":{"title":"FFCV: Accelerating Training by Removing Data Bottlenecks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Aleksander Madry, Andrew Ilyas, Guillaume Leclerc, Hadi Salman, Logan Engstrom, Sung Min Park","submitted_at":"2023-06-21T19:06:41Z","abstract_excerpt":"We present FFCV, a library for easy and fast machine learning model training. FFCV speeds up model training by eliminating (often subtle) data bottlenecks from the training process. In particular, we combine techniques such as an efficient file storage format, caching, data pre-loading, asynchronous data transfer, and just-in-time compilation to (a) make data loading and transfer significantly more efficient, ensuring that GPUs can reach full utilization; and (b) offload as much data processing as possible to the CPU asynchronously, freeing GPU cycles for training. Using FFCV, we train ResNet-"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2306.12517","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-06-21T19:06:41Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"9bc557ecf632ed7309392ed13f1f7710ce032717f3e218cde0658eec180262fa","abstract_canon_sha256":"02cec43b73dbae9e12dd8dee758ae8e8d81c0181b09965c276dc3de4901995ff"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:23:39.541038Z","signature_b64":"azqj1+BJloPqZE1OgVr6NRvENoXO2CmK69y4YBSE+5YOC3A9JZBurk4MOILEtwUY5G15Mh4C0cXh+/P+C1/wAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f090a5f7c5aca9425ccdbdf88c591f4bd58232011e39077204f588f4ec444c29","last_reissued_at":"2026-07-05T06:23:39.540633Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:23:39.540633Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FFCV: Accelerating Training by Removing Data Bottlenecks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Aleksander Madry, Andrew Ilyas, Guillaume Leclerc, Hadi Salman, Logan Engstrom, Sung Min Park","submitted_at":"2023-06-21T19:06:41Z","abstract_excerpt":"We present FFCV, a library for easy and fast machine learning model training. FFCV speeds up model training by eliminating (often subtle) data bottlenecks from the training process. In particular, we combine techniques such as an efficient file storage format, caching, data pre-loading, asynchronous data transfer, and just-in-time compilation to (a) make data loading and transfer significantly more efficient, ensuring that GPUs can reach full utilization; and (b) offload as much data processing as possible to the CPU asynchronously, freeing GPU cycles for training. Using FFCV, we train ResNet-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.12517","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/2306.12517/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2306.12517","created_at":"2026-07-05T06:23:39.540689+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.12517v1","created_at":"2026-07-05T06:23:39.540689+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.12517","created_at":"2026-07-05T06:23:39.540689+00:00"},{"alias_kind":"pith_short_12","alias_value":"6CIKL56FVSUU","created_at":"2026-07-05T06:23:39.540689+00:00"},{"alias_kind":"pith_short_16","alias_value":"6CIKL56FVSUUEXGN","created_at":"2026-07-05T06:23:39.540689+00:00"},{"alias_kind":"pith_short_8","alias_value":"6CIKL56F","created_at":"2026-07-05T06:23:39.540689+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.09112","citing_title":"Mantis Shrimp: Exploring Photometric Band Utilization in Computer Vision Networks for Photometric Redshift Estimation","ref_index":78,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6CIKL56FVSUUEXGNXX4IYWI7JP","json":"https://pith.science/pith/6CIKL56FVSUUEXGNXX4IYWI7JP.json","graph_json":"https://pith.science/api/pith-number/6CIKL56FVSUUEXGNXX4IYWI7JP/graph.json","events_json":"https://pith.science/api/pith-number/6CIKL56FVSUUEXGNXX4IYWI7JP/events.json","paper":"https://pith.science/paper/6CIKL56F"},"agent_actions":{"view_html":"https://pith.science/pith/6CIKL56FVSUUEXGNXX4IYWI7JP","download_json":"https://pith.science/pith/6CIKL56FVSUUEXGNXX4IYWI7JP.json","view_paper":"https://pith.science/paper/6CIKL56F","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.12517&json=true","fetch_graph":"https://pith.science/api/pith-number/6CIKL56FVSUUEXGNXX4IYWI7JP/graph.json","fetch_events":"https://pith.science/api/pith-number/6CIKL56FVSUUEXGNXX4IYWI7JP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6CIKL56FVSUUEXGNXX4IYWI7JP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6CIKL56FVSUUEXGNXX4IYWI7JP/action/storage_attestation","attest_author":"https://pith.science/pith/6CIKL56FVSUUEXGNXX4IYWI7JP/action/author_attestation","sign_citation":"https://pith.science/pith/6CIKL56FVSUUEXGNXX4IYWI7JP/action/citation_signature","submit_replication":"https://pith.science/pith/6CIKL56FVSUUEXGNXX4IYWI7JP/action/replication_record"}},"created_at":"2026-07-05T06:23:39.540689+00:00","updated_at":"2026-07-05T06:23:39.540689+00:00"}