{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:RZRDSNTQNMDYG5V6XLE2IWETMT","short_pith_number":"pith:RZRDSNTQ","schema_version":"1.0","canonical_sha256":"8e623936706b078376bebac9a4589364d61d6581ecf1e1d19c53851cdd96f14f","source":{"kind":"arxiv","id":"2104.14558","version":1},"attestation_state":"computed","paper":{"title":"A Large-Scale Study on Unsupervised Spatiotemporal Representation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Bo Xiong, Christoph Feichtenhofer, Haoqi Fan, Kaiming He, Ross Girshick","submitted_at":"2021-04-29T17:59:53Z","abstract_excerpt":"We present a large-scale study on unsupervised spatiotemporal representation learning from videos. With a unified perspective on four recent image-based frameworks, we study a simple objective that can easily generalize all these methods to space-time. Our objective encourages temporally-persistent features in the same video, and in spite of its simplicity, it works surprisingly well across: (i) different unsupervised frameworks, (ii) pre-training datasets, (iii) downstream datasets, and (iv) backbone architectures. We draw a series of intriguing observations from this study, e.g., we discover"},"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":"2104.14558","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-04-29T17:59:53Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"377fe1837f4e5cd84b0e3c3f1fb0329ccecafad7b3d9ec73321283858796d22c","abstract_canon_sha256":"48379b856daeb37869bd8fa983eac8e6f90aa9489c2aa665d29cdf69ee027eb7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:36:17.862413Z","signature_b64":"kdE0GMDykHbPW773KMxRUDpBe47ohtbWLtW/OjzqGTRFcsu547hiLm81VUmJuTWUa2MffouyS6tffIJd841hCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8e623936706b078376bebac9a4589364d61d6581ecf1e1d19c53851cdd96f14f","last_reissued_at":"2026-07-05T02:36:17.862021Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:36:17.862021Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Large-Scale Study on Unsupervised Spatiotemporal Representation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Bo Xiong, Christoph Feichtenhofer, Haoqi Fan, Kaiming He, Ross Girshick","submitted_at":"2021-04-29T17:59:53Z","abstract_excerpt":"We present a large-scale study on unsupervised spatiotemporal representation learning from videos. With a unified perspective on four recent image-based frameworks, we study a simple objective that can easily generalize all these methods to space-time. Our objective encourages temporally-persistent features in the same video, and in spite of its simplicity, it works surprisingly well across: (i) different unsupervised frameworks, (ii) pre-training datasets, (iii) downstream datasets, and (iv) backbone architectures. We draw a series of intriguing observations from this study, e.g., we discover"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.14558","kind":"arxiv","version":1},"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/2104.14558/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":"2104.14558","created_at":"2026-07-05T02:36:17.862072+00:00"},{"alias_kind":"arxiv_version","alias_value":"2104.14558v1","created_at":"2026-07-05T02:36:17.862072+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.14558","created_at":"2026-07-05T02:36:17.862072+00:00"},{"alias_kind":"pith_short_12","alias_value":"RZRDSNTQNMDY","created_at":"2026-07-05T02:36:17.862072+00:00"},{"alias_kind":"pith_short_16","alias_value":"RZRDSNTQNMDYG5V6","created_at":"2026-07-05T02:36:17.862072+00:00"},{"alias_kind":"pith_short_8","alias_value":"RZRDSNTQ","created_at":"2026-07-05T02:36:17.862072+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.05543","citing_title":"FRAME: Pre-Training Video Feature Representations via Anticipation and Memory","ref_index":13,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RZRDSNTQNMDYG5V6XLE2IWETMT","json":"https://pith.science/pith/RZRDSNTQNMDYG5V6XLE2IWETMT.json","graph_json":"https://pith.science/api/pith-number/RZRDSNTQNMDYG5V6XLE2IWETMT/graph.json","events_json":"https://pith.science/api/pith-number/RZRDSNTQNMDYG5V6XLE2IWETMT/events.json","paper":"https://pith.science/paper/RZRDSNTQ"},"agent_actions":{"view_html":"https://pith.science/pith/RZRDSNTQNMDYG5V6XLE2IWETMT","download_json":"https://pith.science/pith/RZRDSNTQNMDYG5V6XLE2IWETMT.json","view_paper":"https://pith.science/paper/RZRDSNTQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2104.14558&json=true","fetch_graph":"https://pith.science/api/pith-number/RZRDSNTQNMDYG5V6XLE2IWETMT/graph.json","fetch_events":"https://pith.science/api/pith-number/RZRDSNTQNMDYG5V6XLE2IWETMT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RZRDSNTQNMDYG5V6XLE2IWETMT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RZRDSNTQNMDYG5V6XLE2IWETMT/action/storage_attestation","attest_author":"https://pith.science/pith/RZRDSNTQNMDYG5V6XLE2IWETMT/action/author_attestation","sign_citation":"https://pith.science/pith/RZRDSNTQNMDYG5V6XLE2IWETMT/action/citation_signature","submit_replication":"https://pith.science/pith/RZRDSNTQNMDYG5V6XLE2IWETMT/action/replication_record"}},"created_at":"2026-07-05T02:36:17.862072+00:00","updated_at":"2026-07-05T02:36:17.862072+00:00"}