{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:XU7TPAA7PWGARLB5CI43XMBIWQ","short_pith_number":"pith:XU7TPAA7","schema_version":"1.0","canonical_sha256":"bd3f37801f7d8c08ac3d1239bbb028b419336912f93c764a44f11f69ca609079","source":{"kind":"arxiv","id":"2207.02970","version":3},"attestation_state":"computed","paper":{"title":"Network Binarization via Contrastive Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Dan Xu, Liqiang Nie, Yan Yan, Yuzhang Shang, Ziliang Zong","submitted_at":"2022-07-06T21:04:53Z","abstract_excerpt":"Neural network binarization accelerates deep models by quantizing their weights and activations into 1-bit. However, there is still a huge performance gap between Binary Neural Networks (BNNs) and their full-precision (FP) counterparts. As the quantization error caused by weights binarization has been reduced in earlier works, the activations binarization becomes the major obstacle for further improvement of the accuracy. BNN characterises a unique and interesting structure, where the binary and latent FP activations exist in the same forward pass (i.e., $\\text{Binarize}(\\mathbf{a}_F) = \\mathb"},"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":"2207.02970","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-07-06T21:04:53Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"d405c5c53e95e0eda232aeca050a3137951b9048f1cb5357363aea149fe1caae","abstract_canon_sha256":"f734e4b920564906aae1ac0ee272e919a9f2a78280483ad4d2cfc028e0b71cd9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:40:55.055212Z","signature_b64":"8zXp0FOqMHdpPAlKowea7nIAcE1IS3oB52FuDh9dmfCjq4Orlh2apFcyW2ik4dKznPvh4IZaArbt9VX51Qi8Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bd3f37801f7d8c08ac3d1239bbb028b419336912f93c764a44f11f69ca609079","last_reissued_at":"2026-07-05T04:40:55.054803Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:40:55.054803Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Network Binarization via Contrastive Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Dan Xu, Liqiang Nie, Yan Yan, Yuzhang Shang, Ziliang Zong","submitted_at":"2022-07-06T21:04:53Z","abstract_excerpt":"Neural network binarization accelerates deep models by quantizing their weights and activations into 1-bit. However, there is still a huge performance gap between Binary Neural Networks (BNNs) and their full-precision (FP) counterparts. As the quantization error caused by weights binarization has been reduced in earlier works, the activations binarization becomes the major obstacle for further improvement of the accuracy. BNN characterises a unique and interesting structure, where the binary and latent FP activations exist in the same forward pass (i.e., $\\text{Binarize}(\\mathbf{a}_F) = \\mathb"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2207.02970","kind":"arxiv","version":3},"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/2207.02970/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":"2207.02970","created_at":"2026-07-05T04:40:55.054848+00:00"},{"alias_kind":"arxiv_version","alias_value":"2207.02970v3","created_at":"2026-07-05T04:40:55.054848+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2207.02970","created_at":"2026-07-05T04:40:55.054848+00:00"},{"alias_kind":"pith_short_12","alias_value":"XU7TPAA7PWGA","created_at":"2026-07-05T04:40:55.054848+00:00"},{"alias_kind":"pith_short_16","alias_value":"XU7TPAA7PWGARLB5","created_at":"2026-07-05T04:40:55.054848+00:00"},{"alias_kind":"pith_short_8","alias_value":"XU7TPAA7","created_at":"2026-07-05T04:40:55.054848+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.03471","citing_title":"Hyperbolic Binary Neural Network","ref_index":47,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XU7TPAA7PWGARLB5CI43XMBIWQ","json":"https://pith.science/pith/XU7TPAA7PWGARLB5CI43XMBIWQ.json","graph_json":"https://pith.science/api/pith-number/XU7TPAA7PWGARLB5CI43XMBIWQ/graph.json","events_json":"https://pith.science/api/pith-number/XU7TPAA7PWGARLB5CI43XMBIWQ/events.json","paper":"https://pith.science/paper/XU7TPAA7"},"agent_actions":{"view_html":"https://pith.science/pith/XU7TPAA7PWGARLB5CI43XMBIWQ","download_json":"https://pith.science/pith/XU7TPAA7PWGARLB5CI43XMBIWQ.json","view_paper":"https://pith.science/paper/XU7TPAA7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2207.02970&json=true","fetch_graph":"https://pith.science/api/pith-number/XU7TPAA7PWGARLB5CI43XMBIWQ/graph.json","fetch_events":"https://pith.science/api/pith-number/XU7TPAA7PWGARLB5CI43XMBIWQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XU7TPAA7PWGARLB5CI43XMBIWQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XU7TPAA7PWGARLB5CI43XMBIWQ/action/storage_attestation","attest_author":"https://pith.science/pith/XU7TPAA7PWGARLB5CI43XMBIWQ/action/author_attestation","sign_citation":"https://pith.science/pith/XU7TPAA7PWGARLB5CI43XMBIWQ/action/citation_signature","submit_replication":"https://pith.science/pith/XU7TPAA7PWGARLB5CI43XMBIWQ/action/replication_record"}},"created_at":"2026-07-05T04:40:55.054848+00:00","updated_at":"2026-07-05T04:40:55.054848+00:00"}