{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:27FVDDJ6AJHXOYGSTABRD2BYM6","short_pith_number":"pith:27FVDDJ6","schema_version":"1.0","canonical_sha256":"d7cb518d3e024f7760d2980311e83867915aecd531ec6a734164b18df8d825d4","source":{"kind":"arxiv","id":"2212.07525","version":2},"attestation_state":"computed","paper":{"title":"Efficient Self-supervised Learning with Contextualized Target Representations for Vision, Speech and Language","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.SD","eess.AS"],"primary_cat":"cs.LG","authors_text":"Alexei Baevski, Arun Babu, Michael Auli, Wei-Ning Hsu","submitted_at":"2022-12-14T22:13:11Z","abstract_excerpt":"Current self-supervised learning algorithms are often modality-specific and require large amounts of computational resources. To address these issues, we increase the training efficiency of data2vec, a learning objective that generalizes across several modalities. We do not encode masked tokens, use a fast convolutional decoder and amortize the effort to build teacher representations. data2vec 2.0 benefits from the rich contextualized target representations introduced in data2vec which enable a fast self-supervised learner. Experiments on ImageNet-1K image classification show that data2vec 2.0"},"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":"2212.07525","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-12-14T22:13:11Z","cross_cats_sorted":["cs.CL","cs.SD","eess.AS"],"title_canon_sha256":"85f883105fbd7a28b2761eace9c000c2fb77e8b83a57f86ddb8033026b9c7c9a","abstract_canon_sha256":"3a8e41d201637ccce821bf1bf3ed3973cd763dd1f676e246fad81399fd27c4bf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:21:05.247226Z","signature_b64":"hCqrLHzWTYza+vi/arzGWbHhbu08JtBfp5NWn2G/4mfGmtJF0m9T4LHDMLv5eFm0Lpa6TP9MdwJ7UL5oMvEYBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d7cb518d3e024f7760d2980311e83867915aecd531ec6a734164b18df8d825d4","last_reissued_at":"2026-07-05T06:21:05.246745Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:21:05.246745Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficient Self-supervised Learning with Contextualized Target Representations for Vision, Speech and Language","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.SD","eess.AS"],"primary_cat":"cs.LG","authors_text":"Alexei Baevski, Arun Babu, Michael Auli, Wei-Ning Hsu","submitted_at":"2022-12-14T22:13:11Z","abstract_excerpt":"Current self-supervised learning algorithms are often modality-specific and require large amounts of computational resources. To address these issues, we increase the training efficiency of data2vec, a learning objective that generalizes across several modalities. We do not encode masked tokens, use a fast convolutional decoder and amortize the effort to build teacher representations. data2vec 2.0 benefits from the rich contextualized target representations introduced in data2vec which enable a fast self-supervised learner. Experiments on ImageNet-1K image classification show that data2vec 2.0"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.07525","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/2212.07525/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":"2212.07525","created_at":"2026-07-05T06:21:05.246805+00:00"},{"alias_kind":"arxiv_version","alias_value":"2212.07525v2","created_at":"2026-07-05T06:21:05.246805+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.07525","created_at":"2026-07-05T06:21:05.246805+00:00"},{"alias_kind":"pith_short_12","alias_value":"27FVDDJ6AJHX","created_at":"2026-07-05T06:21:05.246805+00:00"},{"alias_kind":"pith_short_16","alias_value":"27FVDDJ6AJHXOYGS","created_at":"2026-07-05T06:21:05.246805+00:00"},{"alias_kind":"pith_short_8","alias_value":"27FVDDJ6","created_at":"2026-07-05T06:21:05.246805+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2404.08471","citing_title":"Revisiting Feature Prediction for Learning Visual Representations from Video","ref_index":222,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/27FVDDJ6AJHXOYGSTABRD2BYM6","json":"https://pith.science/pith/27FVDDJ6AJHXOYGSTABRD2BYM6.json","graph_json":"https://pith.science/api/pith-number/27FVDDJ6AJHXOYGSTABRD2BYM6/graph.json","events_json":"https://pith.science/api/pith-number/27FVDDJ6AJHXOYGSTABRD2BYM6/events.json","paper":"https://pith.science/paper/27FVDDJ6"},"agent_actions":{"view_html":"https://pith.science/pith/27FVDDJ6AJHXOYGSTABRD2BYM6","download_json":"https://pith.science/pith/27FVDDJ6AJHXOYGSTABRD2BYM6.json","view_paper":"https://pith.science/paper/27FVDDJ6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2212.07525&json=true","fetch_graph":"https://pith.science/api/pith-number/27FVDDJ6AJHXOYGSTABRD2BYM6/graph.json","fetch_events":"https://pith.science/api/pith-number/27FVDDJ6AJHXOYGSTABRD2BYM6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/27FVDDJ6AJHXOYGSTABRD2BYM6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/27FVDDJ6AJHXOYGSTABRD2BYM6/action/storage_attestation","attest_author":"https://pith.science/pith/27FVDDJ6AJHXOYGSTABRD2BYM6/action/author_attestation","sign_citation":"https://pith.science/pith/27FVDDJ6AJHXOYGSTABRD2BYM6/action/citation_signature","submit_replication":"https://pith.science/pith/27FVDDJ6AJHXOYGSTABRD2BYM6/action/replication_record"}},"created_at":"2026-07-05T06:21:05.246805+00:00","updated_at":"2026-07-05T06:21:05.246805+00:00"}