{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:R3EZTXH4F6VNWOFBON2SBBQHU4","short_pith_number":"pith:R3EZTXH4","schema_version":"1.0","canonical_sha256":"8ec999dcfc2faadb38a17375208607a70d979df2d9451c54f1cf8968371e6baf","source":{"kind":"arxiv","id":"2110.04403","version":1},"attestation_state":"computed","paper":{"title":"Temperature as Uncertainty in Contrastive Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jasmine Bayrooti, Mike Wu, Noah Goodman, Oliver Zhang","submitted_at":"2021-10-08T23:08:30Z","abstract_excerpt":"Contrastive learning has demonstrated great capability to learn representations without annotations, even outperforming supervised baselines. However, it still lacks important properties useful for real-world application, one of which is uncertainty. In this paper, we propose a simple way to generate uncertainty scores for many contrastive methods by re-purposing temperature, a mysterious hyperparameter used for scaling. By observing that temperature controls how sensitive the objective is to specific embedding locations, we aim to learn temperature as an input-dependent variable, treating it "},"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":"2110.04403","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-08T23:08:30Z","cross_cats_sorted":[],"title_canon_sha256":"fe47be86d86d1dd1d9f1a4b0948987ee2b0eb9f3efae02a4e744ff9b02bcd2f7","abstract_canon_sha256":"68e34418aa369dbff9f76c9da2f908b9f06e9d544cb20872a3dd0aa3e44c0c30"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:21:16.436201Z","signature_b64":"JBWvpGjir5yxUMziesisUVz8LsVX/M/Bp1QVV7AmUoTGjrvLrlPVmTPuvSXaUyXdph3Z3NO16TdZqzVaD14ZBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8ec999dcfc2faadb38a17375208607a70d979df2d9451c54f1cf8968371e6baf","last_reissued_at":"2026-07-05T03:21:16.435639Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:21:16.435639Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Temperature as Uncertainty in Contrastive Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jasmine Bayrooti, Mike Wu, Noah Goodman, Oliver Zhang","submitted_at":"2021-10-08T23:08:30Z","abstract_excerpt":"Contrastive learning has demonstrated great capability to learn representations without annotations, even outperforming supervised baselines. However, it still lacks important properties useful for real-world application, one of which is uncertainty. In this paper, we propose a simple way to generate uncertainty scores for many contrastive methods by re-purposing temperature, a mysterious hyperparameter used for scaling. By observing that temperature controls how sensitive the objective is to specific embedding locations, we aim to learn temperature as an input-dependent variable, treating it "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.04403","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/2110.04403/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":"2110.04403","created_at":"2026-07-05T03:21:16.435699+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.04403v1","created_at":"2026-07-05T03:21:16.435699+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.04403","created_at":"2026-07-05T03:21:16.435699+00:00"},{"alias_kind":"pith_short_12","alias_value":"R3EZTXH4F6VN","created_at":"2026-07-05T03:21:16.435699+00:00"},{"alias_kind":"pith_short_16","alias_value":"R3EZTXH4F6VNWOFB","created_at":"2026-07-05T03:21:16.435699+00:00"},{"alias_kind":"pith_short_8","alias_value":"R3EZTXH4","created_at":"2026-07-05T03:21:16.435699+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.16743","citing_title":"Denoising-While-Completing Network (DWCNet): Robust Point Cloud Completion Under Corruption","ref_index":48,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/R3EZTXH4F6VNWOFBON2SBBQHU4","json":"https://pith.science/pith/R3EZTXH4F6VNWOFBON2SBBQHU4.json","graph_json":"https://pith.science/api/pith-number/R3EZTXH4F6VNWOFBON2SBBQHU4/graph.json","events_json":"https://pith.science/api/pith-number/R3EZTXH4F6VNWOFBON2SBBQHU4/events.json","paper":"https://pith.science/paper/R3EZTXH4"},"agent_actions":{"view_html":"https://pith.science/pith/R3EZTXH4F6VNWOFBON2SBBQHU4","download_json":"https://pith.science/pith/R3EZTXH4F6VNWOFBON2SBBQHU4.json","view_paper":"https://pith.science/paper/R3EZTXH4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.04403&json=true","fetch_graph":"https://pith.science/api/pith-number/R3EZTXH4F6VNWOFBON2SBBQHU4/graph.json","fetch_events":"https://pith.science/api/pith-number/R3EZTXH4F6VNWOFBON2SBBQHU4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/R3EZTXH4F6VNWOFBON2SBBQHU4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/R3EZTXH4F6VNWOFBON2SBBQHU4/action/storage_attestation","attest_author":"https://pith.science/pith/R3EZTXH4F6VNWOFBON2SBBQHU4/action/author_attestation","sign_citation":"https://pith.science/pith/R3EZTXH4F6VNWOFBON2SBBQHU4/action/citation_signature","submit_replication":"https://pith.science/pith/R3EZTXH4F6VNWOFBON2SBBQHU4/action/replication_record"}},"created_at":"2026-07-05T03:21:16.435699+00:00","updated_at":"2026-07-05T03:21:16.435699+00:00"}