{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:QNZUFCNFYOYTZPP6RAZZN6E46B","short_pith_number":"pith:QNZUFCNF","schema_version":"1.0","canonical_sha256":"83734289a5c3b13cbdfe883396f89cf05f3751b82b408ee08fe2a4eee75e411c","source":{"kind":"arxiv","id":"2307.05657","version":2},"attestation_state":"computed","paper":{"title":"Mixed-Precision Quantization for Deep Vision Models with Integer Quadratic Programming","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.NE","authors_text":"Michael Orshansky, Sayeh Sharify, Xin Wang, Zihao Deng","submitted_at":"2023-07-11T15:56:00Z","abstract_excerpt":"Quantization is a widely used technique to compress neural networks. Assigning uniform bit-widths across all layers can result in significant accuracy degradation at low precision and inefficiency at high precision. Mixed-precision quantization (MPQ) addresses this by assigning varied bit-widths to layers, optimizing the accuracy-efficiency trade-off. Existing sensitivity-based methods for MPQ assume that quantization errors across layers are independent, which leads to suboptimal choices. We introduce CLADO, a practical sensitivity-based MPQ algorithm that captures cross-layer dependency of q"},"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":"2307.05657","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.NE","submitted_at":"2023-07-11T15:56:00Z","cross_cats_sorted":[],"title_canon_sha256":"17b995899a839568da2e8085de6eae227502ad97edf87bf232fded4c6dd8d4cc","abstract_canon_sha256":"b65b90264f655de816038743948abdcd5a4d1aa210c70619d68305b2cf567e5f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:45:49.626643Z","signature_b64":"LC79PULRSkiLl5KlzusNMu3zboQ+DRobBfEXyogZiGt4NGA0EJ2j9ENrdHVV+c5nZ5yMldtW9Ka1sARX5a5yDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"83734289a5c3b13cbdfe883396f89cf05f3751b82b408ee08fe2a4eee75e411c","last_reissued_at":"2026-07-05T10:45:49.626224Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:45:49.626224Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Mixed-Precision Quantization for Deep Vision Models with Integer Quadratic Programming","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.NE","authors_text":"Michael Orshansky, Sayeh Sharify, Xin Wang, Zihao Deng","submitted_at":"2023-07-11T15:56:00Z","abstract_excerpt":"Quantization is a widely used technique to compress neural networks. Assigning uniform bit-widths across all layers can result in significant accuracy degradation at low precision and inefficiency at high precision. Mixed-precision quantization (MPQ) addresses this by assigning varied bit-widths to layers, optimizing the accuracy-efficiency trade-off. Existing sensitivity-based methods for MPQ assume that quantization errors across layers are independent, which leads to suboptimal choices. We introduce CLADO, a practical sensitivity-based MPQ algorithm that captures cross-layer dependency of q"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2307.05657","kind":"arxiv","version":2},"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/2307.05657/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":"2307.05657","created_at":"2026-07-05T10:45:49.626280+00:00"},{"alias_kind":"arxiv_version","alias_value":"2307.05657v2","created_at":"2026-07-05T10:45:49.626280+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.05657","created_at":"2026-07-05T10:45:49.626280+00:00"},{"alias_kind":"pith_short_12","alias_value":"QNZUFCNFYOYT","created_at":"2026-07-05T10:45:49.626280+00:00"},{"alias_kind":"pith_short_16","alias_value":"QNZUFCNFYOYTZPP6","created_at":"2026-07-05T10:45:49.626280+00:00"},{"alias_kind":"pith_short_8","alias_value":"QNZUFCNF","created_at":"2026-07-05T10:45:49.626280+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.04861","citing_title":"Mix-QSAM: Mixed-Precision Quantization of the Segment Anything Model","ref_index":6,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QNZUFCNFYOYTZPP6RAZZN6E46B","json":"https://pith.science/pith/QNZUFCNFYOYTZPP6RAZZN6E46B.json","graph_json":"https://pith.science/api/pith-number/QNZUFCNFYOYTZPP6RAZZN6E46B/graph.json","events_json":"https://pith.science/api/pith-number/QNZUFCNFYOYTZPP6RAZZN6E46B/events.json","paper":"https://pith.science/paper/QNZUFCNF"},"agent_actions":{"view_html":"https://pith.science/pith/QNZUFCNFYOYTZPP6RAZZN6E46B","download_json":"https://pith.science/pith/QNZUFCNFYOYTZPP6RAZZN6E46B.json","view_paper":"https://pith.science/paper/QNZUFCNF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2307.05657&json=true","fetch_graph":"https://pith.science/api/pith-number/QNZUFCNFYOYTZPP6RAZZN6E46B/graph.json","fetch_events":"https://pith.science/api/pith-number/QNZUFCNFYOYTZPP6RAZZN6E46B/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QNZUFCNFYOYTZPP6RAZZN6E46B/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QNZUFCNFYOYTZPP6RAZZN6E46B/action/storage_attestation","attest_author":"https://pith.science/pith/QNZUFCNFYOYTZPP6RAZZN6E46B/action/author_attestation","sign_citation":"https://pith.science/pith/QNZUFCNFYOYTZPP6RAZZN6E46B/action/citation_signature","submit_replication":"https://pith.science/pith/QNZUFCNFYOYTZPP6RAZZN6E46B/action/replication_record"}},"created_at":"2026-07-05T10:45:49.626280+00:00","updated_at":"2026-07-05T10:45:49.626280+00:00"}