{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:SHD45MHI7XFI4R7PQXPEWJF4AR","short_pith_number":"pith:SHD45MHI","schema_version":"1.0","canonical_sha256":"91c7ceb0e8fdca8e47ef85de4b24bc0471f774e486c372197871b7e828190fe3","source":{"kind":"arxiv","id":"2501.17683","version":1},"attestation_state":"computed","paper":{"title":"Temperature-Free Loss Function for Contrastive Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bum Jun Kim, Sang Woo Kim","submitted_at":"2025-01-29T14:43:21Z","abstract_excerpt":"As one of the most promising methods in self-supervised learning, contrastive learning has achieved a series of breakthroughs across numerous fields. A predominant approach to implementing contrastive learning is applying InfoNCE loss: By capturing the similarities between pairs, InfoNCE loss enables learning the representation of data. Albeit its success, adopting InfoNCE loss requires tuning a temperature, which is a core hyperparameter for calibrating similarity scores. Despite its significance and sensitivity to performance being emphasized by several studies, searching for a valid tempera"},"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":"2501.17683","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-29T14:43:21Z","cross_cats_sorted":[],"title_canon_sha256":"470943b232d59609f6d1952a840a656d303447d6c4e80f2ff867e904c58ae9c5","abstract_canon_sha256":"c58c15625c2bcead32b55620902c767c1fa10c08262b131337ded507bffefb57"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:06:56.605735Z","signature_b64":"MVcX6QQq0HG7tSx6SN11hzPIswz5LqM1tC5bBpVkxh/yLPo1vRC96tLsVJ8RDq89IDfAaBEC+263UlTzJT8HBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"91c7ceb0e8fdca8e47ef85de4b24bc0471f774e486c372197871b7e828190fe3","last_reissued_at":"2026-07-05T10:06:56.605320Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:06:56.605320Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Temperature-Free Loss Function for Contrastive Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bum Jun Kim, Sang Woo Kim","submitted_at":"2025-01-29T14:43:21Z","abstract_excerpt":"As one of the most promising methods in self-supervised learning, contrastive learning has achieved a series of breakthroughs across numerous fields. A predominant approach to implementing contrastive learning is applying InfoNCE loss: By capturing the similarities between pairs, InfoNCE loss enables learning the representation of data. Albeit its success, adopting InfoNCE loss requires tuning a temperature, which is a core hyperparameter for calibrating similarity scores. Despite its significance and sensitivity to performance being emphasized by several studies, searching for a valid tempera"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.17683","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/2501.17683/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":"2501.17683","created_at":"2026-07-05T10:06:56.605376+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.17683v1","created_at":"2026-07-05T10:06:56.605376+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.17683","created_at":"2026-07-05T10:06:56.605376+00:00"},{"alias_kind":"pith_short_12","alias_value":"SHD45MHI7XFI","created_at":"2026-07-05T10:06:56.605376+00:00"},{"alias_kind":"pith_short_16","alias_value":"SHD45MHI7XFI4R7P","created_at":"2026-07-05T10:06:56.605376+00:00"},{"alias_kind":"pith_short_8","alias_value":"SHD45MHI","created_at":"2026-07-05T10:06:56.605376+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.05068","citing_title":"MaCo-GAN: Manifold-Contrastive Adversarial Learning for Single Image Super-Resolution","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00262","citing_title":"When Softmax Fails at the Top: Extreme Value Corrections for InfoNCE","ref_index":20,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SHD45MHI7XFI4R7PQXPEWJF4AR","json":"https://pith.science/pith/SHD45MHI7XFI4R7PQXPEWJF4AR.json","graph_json":"https://pith.science/api/pith-number/SHD45MHI7XFI4R7PQXPEWJF4AR/graph.json","events_json":"https://pith.science/api/pith-number/SHD45MHI7XFI4R7PQXPEWJF4AR/events.json","paper":"https://pith.science/paper/SHD45MHI"},"agent_actions":{"view_html":"https://pith.science/pith/SHD45MHI7XFI4R7PQXPEWJF4AR","download_json":"https://pith.science/pith/SHD45MHI7XFI4R7PQXPEWJF4AR.json","view_paper":"https://pith.science/paper/SHD45MHI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.17683&json=true","fetch_graph":"https://pith.science/api/pith-number/SHD45MHI7XFI4R7PQXPEWJF4AR/graph.json","fetch_events":"https://pith.science/api/pith-number/SHD45MHI7XFI4R7PQXPEWJF4AR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SHD45MHI7XFI4R7PQXPEWJF4AR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SHD45MHI7XFI4R7PQXPEWJF4AR/action/storage_attestation","attest_author":"https://pith.science/pith/SHD45MHI7XFI4R7PQXPEWJF4AR/action/author_attestation","sign_citation":"https://pith.science/pith/SHD45MHI7XFI4R7PQXPEWJF4AR/action/citation_signature","submit_replication":"https://pith.science/pith/SHD45MHI7XFI4R7PQXPEWJF4AR/action/replication_record"}},"created_at":"2026-07-05T10:06:56.605376+00:00","updated_at":"2026-07-05T10:06:56.605376+00:00"}