{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:HW2J6JQDR3D3Q7IFJDXXPRMSJK","short_pith_number":"pith:HW2J6JQD","schema_version":"1.0","canonical_sha256":"3db49f26038ec7b87d0548ef77c5924a81c95ce04c97c05cb59f79925c56e37d","source":{"kind":"arxiv","id":"2203.15702","version":1},"attestation_state":"computed","paper":{"title":"Contrasting the landscape of contrastive and non-contrastive learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Andrej Risteski, Ashwini Pokle, Jinjin Tian, Yuchen Li","submitted_at":"2022-03-29T16:08:31Z","abstract_excerpt":"A lot of recent advances in unsupervised feature learning are based on designing features which are invariant under semantic data augmentations. A common way to do this is contrastive learning, which uses positive and negative samples. Some recent works however have shown promising results for non-contrastive learning, which does not require negative samples. However, the non-contrastive losses have obvious \"collapsed\" minima, in which the encoders output a constant feature embedding, independent of the input. A folk conjecture is that so long as these collapsed solutions are avoided, the prod"},"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":"2203.15702","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-03-29T16:08:31Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"1df1c483add3b1641b0ea7325ed32499a9b51b355fee2dfdf576cc7884b14bbd","abstract_canon_sha256":"4ba98c1e0b7b4aae73c5a5be136298b47af49f116dd82dc8f8b0cb070c0ba89b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:09:29.468153Z","signature_b64":"rc6wLs+Wy61n5T9Kl2E/d2Ehi5H/YMLkXqkEc8//cntd3y7MAT5/2YfbgqmtPNciDeC9eBO4f9RTSYD1kvt0Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3db49f26038ec7b87d0548ef77c5924a81c95ce04c97c05cb59f79925c56e37d","last_reissued_at":"2026-07-05T04:09:29.467664Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:09:29.467664Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Contrasting the landscape of contrastive and non-contrastive learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Andrej Risteski, Ashwini Pokle, Jinjin Tian, Yuchen Li","submitted_at":"2022-03-29T16:08:31Z","abstract_excerpt":"A lot of recent advances in unsupervised feature learning are based on designing features which are invariant under semantic data augmentations. A common way to do this is contrastive learning, which uses positive and negative samples. Some recent works however have shown promising results for non-contrastive learning, which does not require negative samples. However, the non-contrastive losses have obvious \"collapsed\" minima, in which the encoders output a constant feature embedding, independent of the input. A folk conjecture is that so long as these collapsed solutions are avoided, the prod"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.15702","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/2203.15702/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":"2203.15702","created_at":"2026-07-05T04:09:29.467727+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.15702v1","created_at":"2026-07-05T04:09:29.467727+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.15702","created_at":"2026-07-05T04:09:29.467727+00:00"},{"alias_kind":"pith_short_12","alias_value":"HW2J6JQDR3D3","created_at":"2026-07-05T04:09:29.467727+00:00"},{"alias_kind":"pith_short_16","alias_value":"HW2J6JQDR3D3Q7IF","created_at":"2026-07-05T04:09:29.467727+00:00"},{"alias_kind":"pith_short_8","alias_value":"HW2J6JQD","created_at":"2026-07-05T04:09:29.467727+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22873","citing_title":"SingGuard: A Policy-Adaptive Multimodal LLM Guardrail with Dynamic Reasoning","ref_index":108,"is_internal_anchor":false},{"citing_arxiv_id":"2606.22873","citing_title":"SingGuard: A Policy-Adaptive Multimodal LLM Guardrail with Dynamic Reasoning","ref_index":108,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HW2J6JQDR3D3Q7IFJDXXPRMSJK","json":"https://pith.science/pith/HW2J6JQDR3D3Q7IFJDXXPRMSJK.json","graph_json":"https://pith.science/api/pith-number/HW2J6JQDR3D3Q7IFJDXXPRMSJK/graph.json","events_json":"https://pith.science/api/pith-number/HW2J6JQDR3D3Q7IFJDXXPRMSJK/events.json","paper":"https://pith.science/paper/HW2J6JQD"},"agent_actions":{"view_html":"https://pith.science/pith/HW2J6JQDR3D3Q7IFJDXXPRMSJK","download_json":"https://pith.science/pith/HW2J6JQDR3D3Q7IFJDXXPRMSJK.json","view_paper":"https://pith.science/paper/HW2J6JQD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.15702&json=true","fetch_graph":"https://pith.science/api/pith-number/HW2J6JQDR3D3Q7IFJDXXPRMSJK/graph.json","fetch_events":"https://pith.science/api/pith-number/HW2J6JQDR3D3Q7IFJDXXPRMSJK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HW2J6JQDR3D3Q7IFJDXXPRMSJK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HW2J6JQDR3D3Q7IFJDXXPRMSJK/action/storage_attestation","attest_author":"https://pith.science/pith/HW2J6JQDR3D3Q7IFJDXXPRMSJK/action/author_attestation","sign_citation":"https://pith.science/pith/HW2J6JQDR3D3Q7IFJDXXPRMSJK/action/citation_signature","submit_replication":"https://pith.science/pith/HW2J6JQDR3D3Q7IFJDXXPRMSJK/action/replication_record"}},"created_at":"2026-07-05T04:09:29.467727+00:00","updated_at":"2026-07-05T04:09:29.467727+00:00"}