{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:EQBUFPRQF7DMLA3OAO3KVIGEVE","short_pith_number":"pith:EQBUFPRQ","schema_version":"1.0","canonical_sha256":"240342be302fc6c5836e03b6aaa0c4a9167971d618a4d7a64a90929be4fd9cea","source":{"kind":"arxiv","id":"2112.12750","version":1},"attestation_state":"computed","paper":{"title":"SLIP: Self-supervision meets Language-Image Pre-training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Alexander Kirillov, David Wagner, Norman Mu, Saining Xie","submitted_at":"2021-12-23T18:07:13Z","abstract_excerpt":"Recent work has shown that self-supervised pre-training leads to improvements over supervised learning on challenging visual recognition tasks. CLIP, an exciting new approach to learning with language supervision, demonstrates promising performance on a wide variety of benchmarks. In this work, we explore whether self-supervised learning can aid in the use of language supervision for visual representation learning. We introduce SLIP, a multi-task learning framework for combining self-supervised learning and CLIP pre-training. After pre-training with Vision Transformers, we thoroughly evaluate "},"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":"2112.12750","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-12-23T18:07:13Z","cross_cats_sorted":[],"title_canon_sha256":"d2ca4c776427ff04cd312090f27241da14044151689eab305e09f5eaef78029a","abstract_canon_sha256":"52b5aa7bc550871447be632381f8521020c0ef7816058d958c1e15f407377d12"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:43:30.899168Z","signature_b64":"xYCPlUV6PW/e9feZELpKQhV7mtKz3MRrwwPAL1AA4hR7m0VppERL3KYIyD4r7xZQZgFFod+ZmHw+yG/14ZRQAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"240342be302fc6c5836e03b6aaa0c4a9167971d618a4d7a64a90929be4fd9cea","last_reissued_at":"2026-07-05T03:43:30.898759Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:43:30.898759Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SLIP: Self-supervision meets Language-Image Pre-training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Alexander Kirillov, David Wagner, Norman Mu, Saining Xie","submitted_at":"2021-12-23T18:07:13Z","abstract_excerpt":"Recent work has shown that self-supervised pre-training leads to improvements over supervised learning on challenging visual recognition tasks. CLIP, an exciting new approach to learning with language supervision, demonstrates promising performance on a wide variety of benchmarks. In this work, we explore whether self-supervised learning can aid in the use of language supervision for visual representation learning. We introduce SLIP, a multi-task learning framework for combining self-supervised learning and CLIP pre-training. After pre-training with Vision Transformers, we thoroughly evaluate "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.12750","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/2112.12750/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":"2112.12750","created_at":"2026-07-05T03:43:30.898814+00:00"},{"alias_kind":"arxiv_version","alias_value":"2112.12750v1","created_at":"2026-07-05T03:43:30.898814+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.12750","created_at":"2026-07-05T03:43:30.898814+00:00"},{"alias_kind":"pith_short_12","alias_value":"EQBUFPRQF7DM","created_at":"2026-07-05T03:43:30.898814+00:00"},{"alias_kind":"pith_short_16","alias_value":"EQBUFPRQF7DMLA3O","created_at":"2026-07-05T03:43:30.898814+00:00"},{"alias_kind":"pith_short_8","alias_value":"EQBUFPRQ","created_at":"2026-07-05T03:43:30.898814+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2208.14649","citing_title":"DetailCLIP: Injecting Image Details into CLIP's Feature Space","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2309.16671","citing_title":"Demystifying CLIP Data","ref_index":79,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11095","citing_title":"Bottleneck Tokens for Unified Multimodal Retrieval","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2204.06125","citing_title":"Hierarchical Text-Conditional Image Generation with CLIP Latents","ref_index":31,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EQBUFPRQF7DMLA3OAO3KVIGEVE","json":"https://pith.science/pith/EQBUFPRQF7DMLA3OAO3KVIGEVE.json","graph_json":"https://pith.science/api/pith-number/EQBUFPRQF7DMLA3OAO3KVIGEVE/graph.json","events_json":"https://pith.science/api/pith-number/EQBUFPRQF7DMLA3OAO3KVIGEVE/events.json","paper":"https://pith.science/paper/EQBUFPRQ"},"agent_actions":{"view_html":"https://pith.science/pith/EQBUFPRQF7DMLA3OAO3KVIGEVE","download_json":"https://pith.science/pith/EQBUFPRQF7DMLA3OAO3KVIGEVE.json","view_paper":"https://pith.science/paper/EQBUFPRQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2112.12750&json=true","fetch_graph":"https://pith.science/api/pith-number/EQBUFPRQF7DMLA3OAO3KVIGEVE/graph.json","fetch_events":"https://pith.science/api/pith-number/EQBUFPRQF7DMLA3OAO3KVIGEVE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EQBUFPRQF7DMLA3OAO3KVIGEVE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EQBUFPRQF7DMLA3OAO3KVIGEVE/action/storage_attestation","attest_author":"https://pith.science/pith/EQBUFPRQF7DMLA3OAO3KVIGEVE/action/author_attestation","sign_citation":"https://pith.science/pith/EQBUFPRQF7DMLA3OAO3KVIGEVE/action/citation_signature","submit_replication":"https://pith.science/pith/EQBUFPRQF7DMLA3OAO3KVIGEVE/action/replication_record"}},"created_at":"2026-07-05T03:43:30.898814+00:00","updated_at":"2026-07-05T03:43:30.898814+00:00"}