{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:EBPGHRPUFGWXM3UUUYOBPXOECZ","short_pith_number":"pith:EBPGHRPU","schema_version":"1.0","canonical_sha256":"205e63c5f429ad766e94a61c17ddc416719ad22d0a28fa6dcb9577493ba534b3","source":{"kind":"arxiv","id":"1909.13863","version":1},"attestation_state":"computed","paper":{"title":"XNOR-Net++: Improved Binary Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","eess.IV"],"primary_cat":"cs.CV","authors_text":"Adrian Bulat, Georgios Tzimiropoulos","submitted_at":"2019-09-30T17:42:09Z","abstract_excerpt":"This paper proposes an improved training algorithm for binary neural networks in which both weights and activations are binary numbers. A key but fairly overlooked feature of the current state-of-the-art method of XNOR-Net is the use of analytically calculated real-valued scaling factors for re-weighting the output of binary convolutions. We argue that analytic calculation of these factors is sub-optimal. Instead, in this work, we make the following contributions: (a) we propose to fuse the activation and weight scaling factors into a single one that is learned discriminatively via backpropaga"},"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":"1909.13863","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2019-09-30T17:42:09Z","cross_cats_sorted":["cs.LG","eess.IV"],"title_canon_sha256":"ab3f8160da8835ae52cbb2b8a364aa50d979a73163b02ff497ebecc4e8e86fbb","abstract_canon_sha256":"0cae23c092246f96e5d64a6adc89322a6438478166761f1b59179ed6358e3dc1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:08:14.805756Z","signature_b64":"hGRtIRGtGgyr7Umy+1XYlelRM0DnedCWrJ0dVKXvHVrf7tjLCWfaDml91IQtK0kRZkjUup5nK0dW4AaCudZOAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"205e63c5f429ad766e94a61c17ddc416719ad22d0a28fa6dcb9577493ba534b3","last_reissued_at":"2026-07-05T00:08:14.805348Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:08:14.805348Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"XNOR-Net++: Improved Binary Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","eess.IV"],"primary_cat":"cs.CV","authors_text":"Adrian Bulat, Georgios Tzimiropoulos","submitted_at":"2019-09-30T17:42:09Z","abstract_excerpt":"This paper proposes an improved training algorithm for binary neural networks in which both weights and activations are binary numbers. A key but fairly overlooked feature of the current state-of-the-art method of XNOR-Net is the use of analytically calculated real-valued scaling factors for re-weighting the output of binary convolutions. We argue that analytic calculation of these factors is sub-optimal. Instead, in this work, we make the following contributions: (a) we propose to fuse the activation and weight scaling factors into a single one that is learned discriminatively via backpropaga"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1909.13863","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/1909.13863/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":"1909.13863","created_at":"2026-07-05T00:08:14.805413+00:00"},{"alias_kind":"arxiv_version","alias_value":"1909.13863v1","created_at":"2026-07-05T00:08:14.805413+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1909.13863","created_at":"2026-07-05T00:08:14.805413+00:00"},{"alias_kind":"pith_short_12","alias_value":"EBPGHRPUFGWX","created_at":"2026-07-05T00:08:14.805413+00:00"},{"alias_kind":"pith_short_16","alias_value":"EBPGHRPUFGWXM3UU","created_at":"2026-07-05T00:08:14.805413+00:00"},{"alias_kind":"pith_short_8","alias_value":"EBPGHRPU","created_at":"2026-07-05T00:08:14.805413+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.04456","citing_title":"BiVM: Accurate Binarized Neural Network for Efficient Video Matting","ref_index":10,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EBPGHRPUFGWXM3UUUYOBPXOECZ","json":"https://pith.science/pith/EBPGHRPUFGWXM3UUUYOBPXOECZ.json","graph_json":"https://pith.science/api/pith-number/EBPGHRPUFGWXM3UUUYOBPXOECZ/graph.json","events_json":"https://pith.science/api/pith-number/EBPGHRPUFGWXM3UUUYOBPXOECZ/events.json","paper":"https://pith.science/paper/EBPGHRPU"},"agent_actions":{"view_html":"https://pith.science/pith/EBPGHRPUFGWXM3UUUYOBPXOECZ","download_json":"https://pith.science/pith/EBPGHRPUFGWXM3UUUYOBPXOECZ.json","view_paper":"https://pith.science/paper/EBPGHRPU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1909.13863&json=true","fetch_graph":"https://pith.science/api/pith-number/EBPGHRPUFGWXM3UUUYOBPXOECZ/graph.json","fetch_events":"https://pith.science/api/pith-number/EBPGHRPUFGWXM3UUUYOBPXOECZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EBPGHRPUFGWXM3UUUYOBPXOECZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EBPGHRPUFGWXM3UUUYOBPXOECZ/action/storage_attestation","attest_author":"https://pith.science/pith/EBPGHRPUFGWXM3UUUYOBPXOECZ/action/author_attestation","sign_citation":"https://pith.science/pith/EBPGHRPUFGWXM3UUUYOBPXOECZ/action/citation_signature","submit_replication":"https://pith.science/pith/EBPGHRPUFGWXM3UUUYOBPXOECZ/action/replication_record"}},"created_at":"2026-07-05T00:08:14.805413+00:00","updated_at":"2026-07-05T00:08:14.805413+00:00"}