{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:BPA2I2DJMG6CSULY4L4QEEPE7I","short_pith_number":"pith:BPA2I2DJ","schema_version":"1.0","canonical_sha256":"0bc1a4686961bc295178e2f90211e4fa177b7d5dbd7d43835ee54ee8477981ca","source":{"kind":"arxiv","id":"2106.09943","version":1},"attestation_state":"computed","paper":{"title":"Investigating the Role of Negatives in Contrastive Representation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Akshay Krishnamurthy, Dipendra Misra, Jordan T. Ash, Surbhi Goel","submitted_at":"2021-06-18T06:44:16Z","abstract_excerpt":"Noise contrastive learning is a popular technique for unsupervised representation learning. In this approach, a representation is obtained via reduction to supervised learning, where given a notion of semantic similarity, the learner tries to distinguish a similar (positive) example from a collection of random (negative) examples. The success of modern contrastive learning pipelines relies on many parameters such as the choice of data augmentation, the number of negative examples, and the batch size; however, there is limited understanding as to how these parameters interact and affect downstr"},"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":"2106.09943","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-06-18T06:44:16Z","cross_cats_sorted":["cs.CL","stat.ML"],"title_canon_sha256":"8bcc36e7afc545af19c8f56d265b1305ef8d6588d7eee27605ccaf7f34a04ca1","abstract_canon_sha256":"a0bc1edf30ae7e3d45edf37d8d26d27de34dc95b4bee4c6e69cd8412abf8f0f3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:50:32.072952Z","signature_b64":"kOyp6vUzSPaTykGihMPhPxoXk904zhhV96JoyCTok//IIIb8dbh9m4pO9ja20v9bBcVyv9NtKxsSXsy1XnMnDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0bc1a4686961bc295178e2f90211e4fa177b7d5dbd7d43835ee54ee8477981ca","last_reissued_at":"2026-07-05T02:50:32.072483Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:50:32.072483Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Investigating the Role of Negatives in Contrastive Representation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Akshay Krishnamurthy, Dipendra Misra, Jordan T. Ash, Surbhi Goel","submitted_at":"2021-06-18T06:44:16Z","abstract_excerpt":"Noise contrastive learning is a popular technique for unsupervised representation learning. In this approach, a representation is obtained via reduction to supervised learning, where given a notion of semantic similarity, the learner tries to distinguish a similar (positive) example from a collection of random (negative) examples. The success of modern contrastive learning pipelines relies on many parameters such as the choice of data augmentation, the number of negative examples, and the batch size; however, there is limited understanding as to how these parameters interact and affect downstr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.09943","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/2106.09943/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":"2106.09943","created_at":"2026-07-05T02:50:32.072544+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.09943v1","created_at":"2026-07-05T02:50:32.072544+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.09943","created_at":"2026-07-05T02:50:32.072544+00:00"},{"alias_kind":"pith_short_12","alias_value":"BPA2I2DJMG6C","created_at":"2026-07-05T02:50:32.072544+00:00"},{"alias_kind":"pith_short_16","alias_value":"BPA2I2DJMG6CSULY","created_at":"2026-07-05T02:50:32.072544+00:00"},{"alias_kind":"pith_short_8","alias_value":"BPA2I2DJ","created_at":"2026-07-05T02:50:32.072544+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.02929","citing_title":"OBSER: Object-Based Sub-Environment Recognition for Zero-Shot Environmental Inference","ref_index":4,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BPA2I2DJMG6CSULY4L4QEEPE7I","json":"https://pith.science/pith/BPA2I2DJMG6CSULY4L4QEEPE7I.json","graph_json":"https://pith.science/api/pith-number/BPA2I2DJMG6CSULY4L4QEEPE7I/graph.json","events_json":"https://pith.science/api/pith-number/BPA2I2DJMG6CSULY4L4QEEPE7I/events.json","paper":"https://pith.science/paper/BPA2I2DJ"},"agent_actions":{"view_html":"https://pith.science/pith/BPA2I2DJMG6CSULY4L4QEEPE7I","download_json":"https://pith.science/pith/BPA2I2DJMG6CSULY4L4QEEPE7I.json","view_paper":"https://pith.science/paper/BPA2I2DJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.09943&json=true","fetch_graph":"https://pith.science/api/pith-number/BPA2I2DJMG6CSULY4L4QEEPE7I/graph.json","fetch_events":"https://pith.science/api/pith-number/BPA2I2DJMG6CSULY4L4QEEPE7I/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BPA2I2DJMG6CSULY4L4QEEPE7I/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BPA2I2DJMG6CSULY4L4QEEPE7I/action/storage_attestation","attest_author":"https://pith.science/pith/BPA2I2DJMG6CSULY4L4QEEPE7I/action/author_attestation","sign_citation":"https://pith.science/pith/BPA2I2DJMG6CSULY4L4QEEPE7I/action/citation_signature","submit_replication":"https://pith.science/pith/BPA2I2DJMG6CSULY4L4QEEPE7I/action/replication_record"}},"created_at":"2026-07-05T02:50:32.072544+00:00","updated_at":"2026-07-05T02:50:32.072544+00:00"}