{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:55LYSTC2XJZIUULWOSSD3UZRB5","short_pith_number":"pith:55LYSTC2","schema_version":"1.0","canonical_sha256":"ef57894c5aba728a517674a43dd3310f731fb0eddcb28ba08594b2335c9d99a4","source":{"kind":"arxiv","id":"2306.00238","version":2},"attestation_state":"computed","paper":{"title":"Bytes Are All You Need: Transformers Operating Directly On File Bytes","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ali Farhadi, Maxwell Horton, Mohammad Rastegari, Sachin Mehta","submitted_at":"2023-05-31T23:18:21Z","abstract_excerpt":"Modern deep learning approaches usually utilize modality-specific processing. For example, the most common deep learning approach to image classification involves decoding image file bytes into an RGB tensor which is passed into a neural network. Instead, we investigate modality-independent representation learning by performing classification directly on file bytes, without the need for decoding files at inference time. This enables models to operate on various modalities without any hand-designed, modality-specific processing. Our model, ByteFormer, improves ImageNet Top-1 classification accu"},"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":"2306.00238","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-05-31T23:18:21Z","cross_cats_sorted":[],"title_canon_sha256":"3dcf686d10b399a74d5fb06520c9495a8dfaad840352f95923ecb76938b13e43","abstract_canon_sha256":"fa52ebb96b2481020f755bb4b8fb88085144119a4f8f7a6a541ebe4a9a0e95d5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:38:33.124862Z","signature_b64":"et63Na6jshjmqBH7PxRIF1pObnkClH1GJuYIwzm+tSzRKp0fsEiFVoVSoceDdccXJp6sJGB1Gm8H+2vITg30AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ef57894c5aba728a517674a43dd3310f731fb0eddcb28ba08594b2335c9d99a4","last_reissued_at":"2026-07-05T08:38:33.124407Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:38:33.124407Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bytes Are All You Need: Transformers Operating Directly On File Bytes","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ali Farhadi, Maxwell Horton, Mohammad Rastegari, Sachin Mehta","submitted_at":"2023-05-31T23:18:21Z","abstract_excerpt":"Modern deep learning approaches usually utilize modality-specific processing. For example, the most common deep learning approach to image classification involves decoding image file bytes into an RGB tensor which is passed into a neural network. Instead, we investigate modality-independent representation learning by performing classification directly on file bytes, without the need for decoding files at inference time. This enables models to operate on various modalities without any hand-designed, modality-specific processing. Our model, ByteFormer, improves ImageNet Top-1 classification accu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.00238","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/2306.00238/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":"2306.00238","created_at":"2026-07-05T08:38:33.124465+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.00238v2","created_at":"2026-07-05T08:38:33.124465+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.00238","created_at":"2026-07-05T08:38:33.124465+00:00"},{"alias_kind":"pith_short_12","alias_value":"55LYSTC2XJZI","created_at":"2026-07-05T08:38:33.124465+00:00"},{"alias_kind":"pith_short_16","alias_value":"55LYSTC2XJZIUULW","created_at":"2026-07-05T08:38:33.124465+00:00"},{"alias_kind":"pith_short_8","alias_value":"55LYSTC2","created_at":"2026-07-05T08:38:33.124465+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.28929","citing_title":"Cybersecurity is the True Frontier for Generative AI Success or Failure","ref_index":66,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/55LYSTC2XJZIUULWOSSD3UZRB5","json":"https://pith.science/pith/55LYSTC2XJZIUULWOSSD3UZRB5.json","graph_json":"https://pith.science/api/pith-number/55LYSTC2XJZIUULWOSSD3UZRB5/graph.json","events_json":"https://pith.science/api/pith-number/55LYSTC2XJZIUULWOSSD3UZRB5/events.json","paper":"https://pith.science/paper/55LYSTC2"},"agent_actions":{"view_html":"https://pith.science/pith/55LYSTC2XJZIUULWOSSD3UZRB5","download_json":"https://pith.science/pith/55LYSTC2XJZIUULWOSSD3UZRB5.json","view_paper":"https://pith.science/paper/55LYSTC2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.00238&json=true","fetch_graph":"https://pith.science/api/pith-number/55LYSTC2XJZIUULWOSSD3UZRB5/graph.json","fetch_events":"https://pith.science/api/pith-number/55LYSTC2XJZIUULWOSSD3UZRB5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/55LYSTC2XJZIUULWOSSD3UZRB5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/55LYSTC2XJZIUULWOSSD3UZRB5/action/storage_attestation","attest_author":"https://pith.science/pith/55LYSTC2XJZIUULWOSSD3UZRB5/action/author_attestation","sign_citation":"https://pith.science/pith/55LYSTC2XJZIUULWOSSD3UZRB5/action/citation_signature","submit_replication":"https://pith.science/pith/55LYSTC2XJZIUULWOSSD3UZRB5/action/replication_record"}},"created_at":"2026-07-05T08:38:33.124465+00:00","updated_at":"2026-07-05T08:38:33.124465+00:00"}