{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:2YRZGGYVXSXIW6WJ3N43AW3OXF","short_pith_number":"pith:2YRZGGYV","schema_version":"1.0","canonical_sha256":"d623931b15bcae8b7ac9db79b05b6eb945900a12d6fdd3de04e7efcb8391881b","source":{"kind":"arxiv","id":"1911.12491","version":1},"attestation_state":"computed","paper":{"title":"QKD: Quantization-aware Knowledge Distillation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Chirag Patel, Jangho Kim, Jinwon Lee, Nojun Kwak, Yash Bhalgat","submitted_at":"2019-11-28T02:27:27Z","abstract_excerpt":"Quantization and Knowledge distillation (KD) methods are widely used to reduce memory and power consumption of deep neural networks (DNNs), especially for resource-constrained edge devices. Although their combination is quite promising to meet these requirements, it may not work as desired. It is mainly because the regularization effect of KD further diminishes the already reduced representation power of a quantized model. To address this short-coming, we propose Quantization-aware Knowledge Distillation (QKD) wherein quantization and KD are care-fully coordinated in three phases. First, Self-"},"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":"1911.12491","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2019-11-28T02:27:27Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"7761756529855b8ddc27b6e9c4cb983bdc55e767240fc9cff9cfdbba9a99d565","abstract_canon_sha256":"ab8d29fec83cf9c52b658d59aec3f1b53f58609ac125c05c5293438308efdc41"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:22:45.860959Z","signature_b64":"HWujC2tsonoGxncAznBbWLFObOXn792RBSjSMMJ3NzEcv73gbeDp7KS8iRr2cElbAM8i565opLADnoki5v68BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d623931b15bcae8b7ac9db79b05b6eb945900a12d6fdd3de04e7efcb8391881b","last_reissued_at":"2026-07-05T00:22:45.860455Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:22:45.860455Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"QKD: Quantization-aware Knowledge Distillation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Chirag Patel, Jangho Kim, Jinwon Lee, Nojun Kwak, Yash Bhalgat","submitted_at":"2019-11-28T02:27:27Z","abstract_excerpt":"Quantization and Knowledge distillation (KD) methods are widely used to reduce memory and power consumption of deep neural networks (DNNs), especially for resource-constrained edge devices. Although their combination is quite promising to meet these requirements, it may not work as desired. It is mainly because the regularization effect of KD further diminishes the already reduced representation power of a quantized model. To address this short-coming, we propose Quantization-aware Knowledge Distillation (QKD) wherein quantization and KD are care-fully coordinated in three phases. First, Self-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1911.12491","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/1911.12491/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":"1911.12491","created_at":"2026-07-05T00:22:45.860513+00:00"},{"alias_kind":"arxiv_version","alias_value":"1911.12491v1","created_at":"2026-07-05T00:22:45.860513+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1911.12491","created_at":"2026-07-05T00:22:45.860513+00:00"},{"alias_kind":"pith_short_12","alias_value":"2YRZGGYVXSXI","created_at":"2026-07-05T00:22:45.860513+00:00"},{"alias_kind":"pith_short_16","alias_value":"2YRZGGYVXSXIW6WJ","created_at":"2026-07-05T00:22:45.860513+00:00"},{"alias_kind":"pith_short_8","alias_value":"2YRZGGYV","created_at":"2026-07-05T00:22:45.860513+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.31456","citing_title":"Zero-Shot Quantization for Object Detectors using Off-the-Shelf Generative Models","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27759","citing_title":"Layerwise Progressive Freezing: A Training Scaffold for Depth-Scalable Binary Networks","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10210","citing_title":"Nano-U: Efficient Terrain Segmentation for Tiny Robot Navigation","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2YRZGGYVXSXIW6WJ3N43AW3OXF","json":"https://pith.science/pith/2YRZGGYVXSXIW6WJ3N43AW3OXF.json","graph_json":"https://pith.science/api/pith-number/2YRZGGYVXSXIW6WJ3N43AW3OXF/graph.json","events_json":"https://pith.science/api/pith-number/2YRZGGYVXSXIW6WJ3N43AW3OXF/events.json","paper":"https://pith.science/paper/2YRZGGYV"},"agent_actions":{"view_html":"https://pith.science/pith/2YRZGGYVXSXIW6WJ3N43AW3OXF","download_json":"https://pith.science/pith/2YRZGGYVXSXIW6WJ3N43AW3OXF.json","view_paper":"https://pith.science/paper/2YRZGGYV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1911.12491&json=true","fetch_graph":"https://pith.science/api/pith-number/2YRZGGYVXSXIW6WJ3N43AW3OXF/graph.json","fetch_events":"https://pith.science/api/pith-number/2YRZGGYVXSXIW6WJ3N43AW3OXF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2YRZGGYVXSXIW6WJ3N43AW3OXF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2YRZGGYVXSXIW6WJ3N43AW3OXF/action/storage_attestation","attest_author":"https://pith.science/pith/2YRZGGYVXSXIW6WJ3N43AW3OXF/action/author_attestation","sign_citation":"https://pith.science/pith/2YRZGGYVXSXIW6WJ3N43AW3OXF/action/citation_signature","submit_replication":"https://pith.science/pith/2YRZGGYVXSXIW6WJ3N43AW3OXF/action/replication_record"}},"created_at":"2026-07-05T00:22:45.860513+00:00","updated_at":"2026-07-05T00:22:45.860513+00:00"}