{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:NOOQFNEFPTNXPN75UIILOY3MRF","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":"ffbf20439883f3c42057cfd2d4c29c9a9271efa0cbff84ddca5704b30792b1d9","cross_cats_sorted":["cs.AI","cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-05-16T22:12:25Z","title_canon_sha256":"85d52a10bb18ea16e3e40e8a4901d7730ced87b8c20664c8e1091339f6a85d78"},"schema_version":"1.0","source":{"id":"2305.09828","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.09828","created_at":"2026-07-05T06:10:54Z"},{"alias_kind":"arxiv_version","alias_value":"2305.09828v1","created_at":"2026-07-05T06:10:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.09828","created_at":"2026-07-05T06:10:54Z"},{"alias_kind":"pith_short_12","alias_value":"NOOQFNEFPTNX","created_at":"2026-07-05T06:10:54Z"},{"alias_kind":"pith_short_16","alias_value":"NOOQFNEFPTNXPN75","created_at":"2026-07-05T06:10:54Z"},{"alias_kind":"pith_short_8","alias_value":"NOOQFNEF","created_at":"2026-07-05T06:10:54Z"}],"graph_snapshots":[{"event_id":"sha256:2e8c4dc177e5d3e1dfb5fe26a7d434ac26fe89c92da2deadf4f5d363cc0b0f80","target":"graph","created_at":"2026-07-05T06:10:54Z","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/2305.09828/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"It is notoriously difficult to train Transformers on small datasets; typically, large pre-trained models are instead used as the starting point. We explore the weights of such pre-trained Transformers (particularly for vision) to attempt to find reasons for this discrepancy. Surprisingly, we find that simply initializing the weights of self-attention layers so that they \"look\" more like their pre-trained counterparts allows us to train vanilla Transformers faster and to higher final accuracies, particularly on vision tasks such as CIFAR-10 and ImageNet classification, where we see gains in acc","authors_text":"Asher Trockman, J. Zico Kolter","cross_cats":["cs.AI","cs.LG","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-05-16T22:12:25Z","title":"Mimetic Initialization of Self-Attention Layers"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.09828","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":"sha256:84e55d72acc699d498a811f85cd049fd8a316e2148a6f2b35d1eed535e217fc0","target":"record","created_at":"2026-07-05T06:10:54Z","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":"ffbf20439883f3c42057cfd2d4c29c9a9271efa0cbff84ddca5704b30792b1d9","cross_cats_sorted":["cs.AI","cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-05-16T22:12:25Z","title_canon_sha256":"85d52a10bb18ea16e3e40e8a4901d7730ced87b8c20664c8e1091339f6a85d78"},"schema_version":"1.0","source":{"id":"2305.09828","kind":"arxiv","version":1}},"canonical_sha256":"6b9d02b4857cdb77b7fda210b7636c8945e190e2fd91b4f398911658a1b23e69","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6b9d02b4857cdb77b7fda210b7636c8945e190e2fd91b4f398911658a1b23e69","first_computed_at":"2026-07-05T06:10:54.297549Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:10:54.297549Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"E/6tqo+jjyPi51SE1kS3GDl8+3SKDGal+7yko3WVb4/hUQQjhfuvwktskBhsRhtZDf5/MAh5iZdJ8FEyKmquCg==","signature_status":"signed_v1","signed_at":"2026-07-05T06:10:54.297962Z","signed_message":"canonical_sha256_bytes"},"source_id":"2305.09828","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:84e55d72acc699d498a811f85cd049fd8a316e2148a6f2b35d1eed535e217fc0","sha256:2e8c4dc177e5d3e1dfb5fe26a7d434ac26fe89c92da2deadf4f5d363cc0b0f80"],"state_sha256":"d83f6a69482a42a7ec7a1d2d82f673b028ce039f6d340fa1627fcf1730fbd9a8"}