{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:YFMVAVKXTL3M6RIBZEULHRBL26","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":"767d237127caf9e5c3eaf62d7755cf4c75a6922e35b3e7e2402f7fe7abd2f191","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-06-03T02:08:03Z","title_canon_sha256":"8104df67f42352939eee2a6dfe0ab9c1fae28617c4d10a2ec3169eaea0764ee5"},"schema_version":"1.0","source":{"id":"2106.01548","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2106.01548","created_at":"2026-07-05T04:04:39Z"},{"alias_kind":"arxiv_version","alias_value":"2106.01548v3","created_at":"2026-07-05T04:04:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.01548","created_at":"2026-07-05T04:04:39Z"},{"alias_kind":"pith_short_12","alias_value":"YFMVAVKXTL3M","created_at":"2026-07-05T04:04:39Z"},{"alias_kind":"pith_short_16","alias_value":"YFMVAVKXTL3M6RIB","created_at":"2026-07-05T04:04:39Z"},{"alias_kind":"pith_short_8","alias_value":"YFMVAVKX","created_at":"2026-07-05T04:04:39Z"}],"graph_snapshots":[{"event_id":"sha256:3a2597751572dac6ac9adacc2168eb68062e821719396e4d679aa6de9c7d5fdf","target":"graph","created_at":"2026-07-05T04:04:39Z","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/2106.01548/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Vision Transformers (ViTs) and MLPs signal further efforts on replacing hand-wired features or inductive biases with general-purpose neural architectures. Existing works empower the models by massive data, such as large-scale pre-training and/or repeated strong data augmentations, and still report optimization-related problems (e.g., sensitivity to initialization and learning rates). Hence, this paper investigates ViTs and MLP-Mixers from the lens of loss geometry, intending to improve the models' data efficiency at training and generalization at inference. Visualization and Hessian reveal ext","authors_text":"Boqing Gong, Cho-Jui Hsieh, Xiangning Chen","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-06-03T02:08:03Z","title":"When Vision Transformers Outperform ResNets without Pre-training or Strong Data Augmentations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.01548","kind":"arxiv","version":3},"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:fcb2139b77a5d8c72ab40f182edcc73b1511399908491c5ee6ee73c96367562a","target":"record","created_at":"2026-07-05T04:04:39Z","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":"767d237127caf9e5c3eaf62d7755cf4c75a6922e35b3e7e2402f7fe7abd2f191","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-06-03T02:08:03Z","title_canon_sha256":"8104df67f42352939eee2a6dfe0ab9c1fae28617c4d10a2ec3169eaea0764ee5"},"schema_version":"1.0","source":{"id":"2106.01548","kind":"arxiv","version":3}},"canonical_sha256":"c1595055579af6cf4501c928b3c42bd7a1a1252c7e6a10087bb3224ac2ad9de4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c1595055579af6cf4501c928b3c42bd7a1a1252c7e6a10087bb3224ac2ad9de4","first_computed_at":"2026-07-05T04:04:39.414689Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:04:39.414689Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"OzTq7y2+TCLCdeAt7UhkYmvHNlZfsMd/juw1EyZ0KOpLnDOlcAIxJEYHz1duU9x9hcMmwQky8xB+Fwv5ut+pCA==","signature_status":"signed_v1","signed_at":"2026-07-05T04:04:39.415955Z","signed_message":"canonical_sha256_bytes"},"source_id":"2106.01548","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:fcb2139b77a5d8c72ab40f182edcc73b1511399908491c5ee6ee73c96367562a","sha256:3a2597751572dac6ac9adacc2168eb68062e821719396e4d679aa6de9c7d5fdf"],"state_sha256":"1229117f25bceb42ef49d6f3aaf33eb17d8b39f5afba45183d45615caa1cdbae"}