{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:VKM6NKHHDFRCQ4RKL4VN7CV4VZ","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":"6238adccafeedeef2db5d7eb1dd165dcc2a780a0709a21bf0621578993f81067","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-11-11T18:46:40Z","title_canon_sha256":"ea6c7158eab8cc332ab8f2aacdb3daf765d105415ff67a4b06ad26fe5613a9a1"},"schema_version":"1.0","source":{"id":"2111.06377","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2111.06377","created_at":"2026-05-17T23:38:48Z"},{"alias_kind":"arxiv_version","alias_value":"2111.06377v3","created_at":"2026-05-17T23:38:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2111.06377","created_at":"2026-05-17T23:38:48Z"},{"alias_kind":"pith_short_12","alias_value":"VKM6NKHHDFRC","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_16","alias_value":"VKM6NKHHDFRCQ4RK","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_8","alias_value":"VKM6NKHH","created_at":"2026-05-18T12:33:33Z"}],"graph_snapshots":[{"event_id":"sha256:f0a7f1e4bd5aef0c2a3655e176d6070c4b237698a4b048ff5c865576aba9de97","target":"graph","created_at":"2026-05-17T23:38:48Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Our scalable approach allows for learning high-capacity models that generalize well: e.g., a vanilla ViT-Huge model achieves the best accuracy (87.8%) among methods that use only ImageNet-1K data."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That masking a high proportion of the input (e.g. 75%) yields a nontrivial and meaningful self-supervisory task whose difficulty drives useful feature learning rather than trivial solutions."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Masked autoencoders with asymmetric encoder-decoder and 75% masking ratio enable scalable self-supervised pre-training of vision transformers, achieving 87.8% ImageNet-1K accuracy with ViT-Huge using only unlabeled data."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Masked autoencoders learn scalable vision features by reconstructing heavily masked image patches."}],"snapshot_sha256":"c817407f7174acfd1ba55553d8686b8d2df559320a3d10e33387caa6b0d6fec7"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"f9f4c7850199158c1e43f43474bd607f847376ab3b62ff8e8839577725c6a4bd"},"paper":{"abstract_excerpt":"This paper shows that masked autoencoders (MAE) are scalable self-supervised learners for computer vision. Our MAE approach is simple: we mask random patches of the input image and reconstruct the missing pixels. It is based on two core designs. First, we develop an asymmetric encoder-decoder architecture, with an encoder that operates only on the visible subset of patches (without mask tokens), along with a lightweight decoder that reconstructs the original image from the latent representation and mask tokens. Second, we find that masking a high proportion of the input image, e.g., 75%, yield","authors_text":"Kaiming He, Piotr Doll\\'ar, Ross Girshick, Saining Xie, Xinlei Chen, Yanghao Li","cross_cats":[],"headline":"Masked autoencoders learn scalable vision features by reconstructing heavily masked image patches.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-11-11T18:46:40Z","title":"Masked Autoencoders Are Scalable Vision Learners"},"references":{"count":74,"internal_anchors":5,"resolved_work":74,"sample":[{"cited_arxiv_id":"1607.06450","doi":"","is_internal_anchor":true,"ref_index":1,"title":"Layer Normalization","work_id":"20a2d720-0046-4c7c-bcd6-327ec8143f69","year":2016},{"cited_arxiv_id":"2106.08254","doi":"","is_internal_anchor":true,"ref_index":2,"title":"BEiT: BERT Pre-Training of Image Transformers","work_id":"d74eda3c-bf7e-45f1-a8f1-a0137ecca3f4","year":2021},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Self-organizing neural network that discovers surfaces in random-dot stereograms","work_id":"2a90a785-1b05-47c2-b9f2-70f36e69df81","year":1992},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Language mod- els are few-shot learners","work_id":"0b82323e-a2a3-43a2-8d5c-6b111825c722","year":2020},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Emerging properties in self-supervised vision transformers","work_id":"b6e4b8f1-b87c-4a46-aa8f-ee63e646489a","year":2021}],"snapshot_sha256":"11d01219b550958a318eef7ab206f724197244f75f5cb537e2353204f99fc962"},"source":{"id":"2111.06377","kind":"arxiv","version":3},"verdict":{"created_at":"2026-05-16T06:49:33.669252Z","id":"7a27ae39-4672-4f49-871f-747c0173bae4","model_set":{"reader":"grok-4.3"},"one_line_summary":"Masked autoencoders with asymmetric encoder-decoder and 75% masking ratio enable scalable self-supervised pre-training of vision transformers, achieving 87.8% ImageNet-1K accuracy with ViT-Huge using only unlabeled data.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Masked autoencoders learn scalable vision features by reconstructing heavily masked image patches.","strongest_claim":"Our scalable approach allows for learning high-capacity models that generalize well: e.g., a vanilla ViT-Huge model achieves the best accuracy (87.8%) among methods that use only ImageNet-1K data.","weakest_assumption":"That masking a high proportion of the input (e.g. 75%) yields a nontrivial and meaningful self-supervisory task whose difficulty drives useful feature learning rather than trivial solutions."}},"verdict_id":"7a27ae39-4672-4f49-871f-747c0173bae4"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:6b6834059ef97eddaa7aaebcbb58ddeb281d720f1a6ed2a7e0a36f38ed240cb5","target":"record","created_at":"2026-05-17T23:38:48Z","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":"6238adccafeedeef2db5d7eb1dd165dcc2a780a0709a21bf0621578993f81067","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-11-11T18:46:40Z","title_canon_sha256":"ea6c7158eab8cc332ab8f2aacdb3daf765d105415ff67a4b06ad26fe5613a9a1"},"schema_version":"1.0","source":{"id":"2111.06377","kind":"arxiv","version":3}},"canonical_sha256":"aa99e6a8e7196228722a5f2adf8abcae7a418c36a55f77187f94a3a23e1c1656","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"aa99e6a8e7196228722a5f2adf8abcae7a418c36a55f77187f94a3a23e1c1656","first_computed_at":"2026-05-17T23:38:48.776710Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:38:48.776710Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"hCA0rE2tKv/dTdXV18kT9aVoOXLfSMGSOvI0A7K3DMjW8zp1kWH2hjdzkQokpxuieX+m7OSW5/DSH95PjOR0BQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:38:48.777379Z","signed_message":"canonical_sha256_bytes"},"source_id":"2111.06377","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6b6834059ef97eddaa7aaebcbb58ddeb281d720f1a6ed2a7e0a36f38ed240cb5","sha256:f0a7f1e4bd5aef0c2a3655e176d6070c4b237698a4b048ff5c865576aba9de97"],"state_sha256":"6b30affc8351da9d22423c8bf7e7b86cab90816e94a90ec9e764270faf5879c6"}