{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:NOMJVEX5RGOO36FJHBNXUMECHZ","short_pith_number":"pith:NOMJVEX5","canonical_record":{"source":{"id":"2205.14204","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-05-27T19:09:42Z","cross_cats_sorted":[],"title_canon_sha256":"da2e20ac9a7e5448dfd968fece8d5c264627903d6ed799c883c78e5ea98f45f5","abstract_canon_sha256":"48bccf1831c40511f24363a1314da4cd81441b220291ab590f356b7a2dcc1b21"},"schema_version":"1.0"},"canonical_sha256":"6b989a92fd899cedf8a9385b7a30823e4892957c17080b74573d29999c7974e3","source":{"kind":"arxiv","id":"2205.14204","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2205.14204","created_at":"2026-07-05T05:08:52Z"},{"alias_kind":"arxiv_version","alias_value":"2205.14204v3","created_at":"2026-07-05T05:08:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.14204","created_at":"2026-07-05T05:08:52Z"},{"alias_kind":"pith_short_12","alias_value":"NOMJVEX5RGOO","created_at":"2026-07-05T05:08:52Z"},{"alias_kind":"pith_short_16","alias_value":"NOMJVEX5RGOO36FJ","created_at":"2026-07-05T05:08:52Z"},{"alias_kind":"pith_short_8","alias_value":"NOMJVEX5","created_at":"2026-07-05T05:08:52Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:NOMJVEX5RGOO36FJHBNXUMECHZ","target":"record","payload":{"canonical_record":{"source":{"id":"2205.14204","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-05-27T19:09:42Z","cross_cats_sorted":[],"title_canon_sha256":"da2e20ac9a7e5448dfd968fece8d5c264627903d6ed799c883c78e5ea98f45f5","abstract_canon_sha256":"48bccf1831c40511f24363a1314da4cd81441b220291ab590f356b7a2dcc1b21"},"schema_version":"1.0"},"canonical_sha256":"6b989a92fd899cedf8a9385b7a30823e4892957c17080b74573d29999c7974e3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:08:52.160749Z","signature_b64":"3wrC8eC0z1mlg3oytqI3BA4qXx97DZg6d6LCKpBImnTVZDPhoSdor4grPPM8DdRDnVEBKhxwWA8PdiP8p+87BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6b989a92fd899cedf8a9385b7a30823e4892957c17080b74573d29999c7974e3","last_reissued_at":"2026-07-05T05:08:52.160334Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:08:52.160334Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2205.14204","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:08:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"16vy8VV9o7Cg8CjCRUEoRh73UIj8CoYxFW3TDObXD9UYOWE0Az7AgAq0MCHZS/ZqAY9xmIruVSxWLfqDNRHzAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T05:36:58.034350Z"},"content_sha256":"be06030df6c3caa659df1eb57627d3014fd6d8968df34f63138b2b0ee2889f16","schema_version":"1.0","event_id":"sha256:be06030df6c3caa659df1eb57627d3014fd6d8968df34f63138b2b0ee2889f16"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:NOMJVEX5RGOO36FJHBNXUMECHZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Multimodal Masked Autoencoders Learn Transferable Representations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dale Schuurmans, Hao Liu, Lisa Lee, Pieter Abbeel, Sergey Levine, Xinyang Geng","submitted_at":"2022-05-27T19:09:42Z","abstract_excerpt":"Building scalable models to learn from diverse, multimodal data remains an open challenge. For vision-language data, the dominant approaches are based on contrastive learning objectives that train a separate encoder for each modality. While effective, contrastive learning approaches introduce sampling bias depending on the data augmentations used, which can degrade performance on downstream tasks. Moreover, these methods are limited to paired image-text data, and cannot leverage widely-available unpaired data. In this paper, we investigate whether a large multimodal model trained purely via ma"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.14204","kind":"arxiv","version":3},"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/2205.14204/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:08:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3WnS7YQcRSNGBTblZUmWMPjSyS/6He+eKINSZ4UVYXC+q5t2Ht5VzEadTAEkeH8jsz8PsMd7fY+/9EXVnCSfBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T05:36:58.034863Z"},"content_sha256":"f9ff40dbce286ff67034509dd892f25ebd24867008927bcab72bfc244056cb8d","schema_version":"1.0","event_id":"sha256:f9ff40dbce286ff67034509dd892f25ebd24867008927bcab72bfc244056cb8d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NOMJVEX5RGOO36FJHBNXUMECHZ/bundle.json","state_url":"https://pith.science/pith/NOMJVEX5RGOO36FJHBNXUMECHZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NOMJVEX5RGOO36FJHBNXUMECHZ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-04T05:36:58Z","links":{"resolver":"https://pith.science/pith/NOMJVEX5RGOO36FJHBNXUMECHZ","bundle":"https://pith.science/pith/NOMJVEX5RGOO36FJHBNXUMECHZ/bundle.json","state":"https://pith.science/pith/NOMJVEX5RGOO36FJHBNXUMECHZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NOMJVEX5RGOO36FJHBNXUMECHZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:NOMJVEX5RGOO36FJHBNXUMECHZ","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":"48bccf1831c40511f24363a1314da4cd81441b220291ab590f356b7a2dcc1b21","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-05-27T19:09:42Z","title_canon_sha256":"da2e20ac9a7e5448dfd968fece8d5c264627903d6ed799c883c78e5ea98f45f5"},"schema_version":"1.0","source":{"id":"2205.14204","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2205.14204","created_at":"2026-07-05T05:08:52Z"},{"alias_kind":"arxiv_version","alias_value":"2205.14204v3","created_at":"2026-07-05T05:08:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.14204","created_at":"2026-07-05T05:08:52Z"},{"alias_kind":"pith_short_12","alias_value":"NOMJVEX5RGOO","created_at":"2026-07-05T05:08:52Z"},{"alias_kind":"pith_short_16","alias_value":"NOMJVEX5RGOO36FJ","created_at":"2026-07-05T05:08:52Z"},{"alias_kind":"pith_short_8","alias_value":"NOMJVEX5","created_at":"2026-07-05T05:08:52Z"}],"graph_snapshots":[{"event_id":"sha256:f9ff40dbce286ff67034509dd892f25ebd24867008927bcab72bfc244056cb8d","target":"graph","created_at":"2026-07-05T05:08:52Z","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/2205.14204/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Building scalable models to learn from diverse, multimodal data remains an open challenge. For vision-language data, the dominant approaches are based on contrastive learning objectives that train a separate encoder for each modality. While effective, contrastive learning approaches introduce sampling bias depending on the data augmentations used, which can degrade performance on downstream tasks. Moreover, these methods are limited to paired image-text data, and cannot leverage widely-available unpaired data. In this paper, we investigate whether a large multimodal model trained purely via ma","authors_text":"Dale Schuurmans, Hao Liu, Lisa Lee, Pieter Abbeel, Sergey Levine, Xinyang Geng","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-05-27T19:09:42Z","title":"Multimodal Masked Autoencoders Learn Transferable Representations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.14204","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:be06030df6c3caa659df1eb57627d3014fd6d8968df34f63138b2b0ee2889f16","target":"record","created_at":"2026-07-05T05:08:52Z","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":"48bccf1831c40511f24363a1314da4cd81441b220291ab590f356b7a2dcc1b21","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-05-27T19:09:42Z","title_canon_sha256":"da2e20ac9a7e5448dfd968fece8d5c264627903d6ed799c883c78e5ea98f45f5"},"schema_version":"1.0","source":{"id":"2205.14204","kind":"arxiv","version":3}},"canonical_sha256":"6b989a92fd899cedf8a9385b7a30823e4892957c17080b74573d29999c7974e3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6b989a92fd899cedf8a9385b7a30823e4892957c17080b74573d29999c7974e3","first_computed_at":"2026-07-05T05:08:52.160334Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:08:52.160334Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3wrC8eC0z1mlg3oytqI3BA4qXx97DZg6d6LCKpBImnTVZDPhoSdor4grPPM8DdRDnVEBKhxwWA8PdiP8p+87BQ==","signature_status":"signed_v1","signed_at":"2026-07-05T05:08:52.160749Z","signed_message":"canonical_sha256_bytes"},"source_id":"2205.14204","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:be06030df6c3caa659df1eb57627d3014fd6d8968df34f63138b2b0ee2889f16","sha256:f9ff40dbce286ff67034509dd892f25ebd24867008927bcab72bfc244056cb8d"],"state_sha256":"7c4d2b4b0f55c897410f21dc26d80f631ef9ba2b208f5dce18c49e3ad51c0479"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Nh2KIyQi1ZXY39ArS7o0hSEmwvOwn/fANZ7tw22Fop1YQ2g7nA5W9uc+SRbG+nMllkUdCSVSMBRS/30Tyk9eCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T05:36:58.038468Z","bundle_sha256":"b27323d16e99f39410d900b6631b33f21baa2831eca559be23c5510efd9cc485"}}