{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:C62ZKZ7MBWGDQCU6AOYWSHNRW4","short_pith_number":"pith:C62ZKZ7M","schema_version":"1.0","canonical_sha256":"17b59567ec0d8c380a9e03b1691db1b7338da620a22aed24c376d8e8456c0c1a","source":{"kind":"arxiv","id":"2308.01140","version":2},"attestation_state":"computed","paper":{"title":"Dynamically Scaled Temperature in Self-Supervised Contrastive Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Rakesh Dey, Saumik Bhattacharya, Siladittya Manna, Soumitri Chattopadhyay, Umapada Pal","submitted_at":"2023-08-02T13:31:41Z","abstract_excerpt":"In contemporary self-supervised contrastive algorithms like SimCLR, MoCo, etc., the task of balancing attraction between two semantically similar samples and repulsion between two samples of different classes is primarily affected by the presence of hard negative samples. While the InfoNCE loss has been shown to impose penalties based on hardness, the temperature hyper-parameter is the key to regulating the penalties and the trade-off between uniformity and tolerance. In this work, we focus our attention on improving the performance of InfoNCE loss in self-supervised learning by proposing a no"},"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":"2308.01140","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-08-02T13:31:41Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"98e63f81946fb983112f5531752e44a6330fbb04b892cb3d46a704307c790da7","abstract_canon_sha256":"499dabe9656ccc9a873446c9d2a088107731157046de48f8db35ff87ade548dc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:17:48.051065Z","signature_b64":"MxD9wLiVCfbEZH3yHU1K61O3SumdR1xczEhkjUt0wHaOmT9nt5+LQNenJPwYKc11y0wKjMDBGMyad3CTPeLXAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"17b59567ec0d8c380a9e03b1691db1b7338da620a22aed24c376d8e8456c0c1a","last_reissued_at":"2026-07-05T08:17:48.050604Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:17:48.050604Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dynamically Scaled Temperature in Self-Supervised Contrastive Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Rakesh Dey, Saumik Bhattacharya, Siladittya Manna, Soumitri Chattopadhyay, Umapada Pal","submitted_at":"2023-08-02T13:31:41Z","abstract_excerpt":"In contemporary self-supervised contrastive algorithms like SimCLR, MoCo, etc., the task of balancing attraction between two semantically similar samples and repulsion between two samples of different classes is primarily affected by the presence of hard negative samples. While the InfoNCE loss has been shown to impose penalties based on hardness, the temperature hyper-parameter is the key to regulating the penalties and the trade-off between uniformity and tolerance. In this work, we focus our attention on improving the performance of InfoNCE loss in self-supervised learning by proposing a no"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.01140","kind":"arxiv","version":2},"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/2308.01140/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":"2308.01140","created_at":"2026-07-05T08:17:48.050659+00:00"},{"alias_kind":"arxiv_version","alias_value":"2308.01140v2","created_at":"2026-07-05T08:17:48.050659+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.01140","created_at":"2026-07-05T08:17:48.050659+00:00"},{"alias_kind":"pith_short_12","alias_value":"C62ZKZ7MBWGD","created_at":"2026-07-05T08:17:48.050659+00:00"},{"alias_kind":"pith_short_16","alias_value":"C62ZKZ7MBWGDQCU6","created_at":"2026-07-05T08:17:48.050659+00:00"},{"alias_kind":"pith_short_8","alias_value":"C62ZKZ7M","created_at":"2026-07-05T08:17:48.050659+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.03077","citing_title":"Self-supervised Radio Representation Learning: Can we Learn Multiple Tasks?","ref_index":2023,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/C62ZKZ7MBWGDQCU6AOYWSHNRW4","json":"https://pith.science/pith/C62ZKZ7MBWGDQCU6AOYWSHNRW4.json","graph_json":"https://pith.science/api/pith-number/C62ZKZ7MBWGDQCU6AOYWSHNRW4/graph.json","events_json":"https://pith.science/api/pith-number/C62ZKZ7MBWGDQCU6AOYWSHNRW4/events.json","paper":"https://pith.science/paper/C62ZKZ7M"},"agent_actions":{"view_html":"https://pith.science/pith/C62ZKZ7MBWGDQCU6AOYWSHNRW4","download_json":"https://pith.science/pith/C62ZKZ7MBWGDQCU6AOYWSHNRW4.json","view_paper":"https://pith.science/paper/C62ZKZ7M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2308.01140&json=true","fetch_graph":"https://pith.science/api/pith-number/C62ZKZ7MBWGDQCU6AOYWSHNRW4/graph.json","fetch_events":"https://pith.science/api/pith-number/C62ZKZ7MBWGDQCU6AOYWSHNRW4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/C62ZKZ7MBWGDQCU6AOYWSHNRW4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/C62ZKZ7MBWGDQCU6AOYWSHNRW4/action/storage_attestation","attest_author":"https://pith.science/pith/C62ZKZ7MBWGDQCU6AOYWSHNRW4/action/author_attestation","sign_citation":"https://pith.science/pith/C62ZKZ7MBWGDQCU6AOYWSHNRW4/action/citation_signature","submit_replication":"https://pith.science/pith/C62ZKZ7MBWGDQCU6AOYWSHNRW4/action/replication_record"}},"created_at":"2026-07-05T08:17:48.050659+00:00","updated_at":"2026-07-05T08:17:48.050659+00:00"}