{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:5ZJW2TX4SRUQHTTPC7DC6DPAX7","short_pith_number":"pith:5ZJW2TX4","schema_version":"1.0","canonical_sha256":"ee536d4efc946903ce6f17c62f0de0bfe09609fa91c94769dc563d721cb235f4","source":{"kind":"arxiv","id":"2104.12081","version":3},"attestation_state":"computed","paper":{"title":"How Well Does Self-Supervised Pre-Training Perform with Streaming Data?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Dapeng Hu, Hailin Hu, Jiashi Feng, Lanqing Hong, Qizhengqiu Lu, Shipeng Yan, Xinchao Wang, Yifan Zhang, Zhenguo Li","submitted_at":"2021-04-25T06:56:48Z","abstract_excerpt":"Prior works on self-supervised pre-training focus on the joint training scenario, where massive unlabeled data are assumed to be given as input all at once, and only then is a learner trained. Unfortunately, such a problem setting is often impractical if not infeasible since many real-world tasks rely on sequential learning, e.g., data are decentralized or collected in a streaming fashion. In this paper, we conduct the first thorough and dedicated investigation on self-supervised pre-training with streaming data, aiming to shed light on the model behavior under this overlooked setup. Specifica"},"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":"2104.12081","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-04-25T06:56:48Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"754fdb1b081cf27f9a6d90c1241a2632f2c060f10f7b224e42b1f7709201f9ef","abstract_canon_sha256":"eafa7b5a6d0716cfbf24127c65824d8ad375c91f0a2ae830f6905da1797b2131"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:42:13.420729Z","signature_b64":"BYYEAHuauk+s+jshCMsI7uX9bqtf9mm/NMpnigoME97FraCQfBk30OEzZd04guos61kpM11UW0tUCpFe/aXdAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ee536d4efc946903ce6f17c62f0de0bfe09609fa91c94769dc563d721cb235f4","last_reissued_at":"2026-07-05T04:42:13.420286Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:42:13.420286Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"How Well Does Self-Supervised Pre-Training Perform with Streaming Data?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Dapeng Hu, Hailin Hu, Jiashi Feng, Lanqing Hong, Qizhengqiu Lu, Shipeng Yan, Xinchao Wang, Yifan Zhang, Zhenguo Li","submitted_at":"2021-04-25T06:56:48Z","abstract_excerpt":"Prior works on self-supervised pre-training focus on the joint training scenario, where massive unlabeled data are assumed to be given as input all at once, and only then is a learner trained. Unfortunately, such a problem setting is often impractical if not infeasible since many real-world tasks rely on sequential learning, e.g., data are decentralized or collected in a streaming fashion. In this paper, we conduct the first thorough and dedicated investigation on self-supervised pre-training with streaming data, aiming to shed light on the model behavior under this overlooked setup. Specifica"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.12081","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/2104.12081/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":"2104.12081","created_at":"2026-07-05T04:42:13.420358+00:00"},{"alias_kind":"arxiv_version","alias_value":"2104.12081v3","created_at":"2026-07-05T04:42:13.420358+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.12081","created_at":"2026-07-05T04:42:13.420358+00:00"},{"alias_kind":"pith_short_12","alias_value":"5ZJW2TX4SRUQ","created_at":"2026-07-05T04:42:13.420358+00:00"},{"alias_kind":"pith_short_16","alias_value":"5ZJW2TX4SRUQHTTP","created_at":"2026-07-05T04:42:13.420358+00:00"},{"alias_kind":"pith_short_8","alias_value":"5ZJW2TX4","created_at":"2026-07-05T04:42:13.420358+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.14364","citing_title":"MoRe: Modular Representations for Principled Continual Representation Learning on Sequential Data","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14364","citing_title":"MoRe: Modular Representations for Principled Continual Representation Learning on Sequential Data","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14364","citing_title":"MoRe: Modular Representations for Principled Continual Representation Learning on Sequential Data","ref_index":9,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5ZJW2TX4SRUQHTTPC7DC6DPAX7","json":"https://pith.science/pith/5ZJW2TX4SRUQHTTPC7DC6DPAX7.json","graph_json":"https://pith.science/api/pith-number/5ZJW2TX4SRUQHTTPC7DC6DPAX7/graph.json","events_json":"https://pith.science/api/pith-number/5ZJW2TX4SRUQHTTPC7DC6DPAX7/events.json","paper":"https://pith.science/paper/5ZJW2TX4"},"agent_actions":{"view_html":"https://pith.science/pith/5ZJW2TX4SRUQHTTPC7DC6DPAX7","download_json":"https://pith.science/pith/5ZJW2TX4SRUQHTTPC7DC6DPAX7.json","view_paper":"https://pith.science/paper/5ZJW2TX4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2104.12081&json=true","fetch_graph":"https://pith.science/api/pith-number/5ZJW2TX4SRUQHTTPC7DC6DPAX7/graph.json","fetch_events":"https://pith.science/api/pith-number/5ZJW2TX4SRUQHTTPC7DC6DPAX7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5ZJW2TX4SRUQHTTPC7DC6DPAX7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5ZJW2TX4SRUQHTTPC7DC6DPAX7/action/storage_attestation","attest_author":"https://pith.science/pith/5ZJW2TX4SRUQHTTPC7DC6DPAX7/action/author_attestation","sign_citation":"https://pith.science/pith/5ZJW2TX4SRUQHTTPC7DC6DPAX7/action/citation_signature","submit_replication":"https://pith.science/pith/5ZJW2TX4SRUQHTTPC7DC6DPAX7/action/replication_record"}},"created_at":"2026-07-05T04:42:13.420358+00:00","updated_at":"2026-07-05T04:42:13.420358+00:00"}