{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:RYMD2RJ5RRSTEYFYYKRBQSMTKV","short_pith_number":"pith:RYMD2RJ5","schema_version":"1.0","canonical_sha256":"8e183d453d8c653260b8c2a21849935547e1f54fab2f2fee0dfdbb771ce6821f","source":{"kind":"arxiv","id":"1712.05877","version":1},"attestation_state":"computed","paper":{"title":"Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Andrew Howard, Benoit Jacob, Bo Chen, Dmitry Kalenichenko, Hartwig Adam, Matthew Tang, Menglong Zhu, Skirmantas Kligys","submitted_at":"2017-12-15T23:56:52Z","abstract_excerpt":"The rising popularity of intelligent mobile devices and the daunting computational cost of deep learning-based models call for efficient and accurate on-device inference schemes. We propose a quantization scheme that allows inference to be carried out using integer-only arithmetic, which can be implemented more efficiently than floating point inference on commonly available integer-only hardware. We also co-design a training procedure to preserve end-to-end model accuracy post quantization. As a result, the proposed quantization scheme improves the tradeoff between accuracy and on-device laten"},"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":"1712.05877","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-12-15T23:56:52Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"317703ba0e36dc104e23217e9e42a7ebd06ddf30d47dc2d056e1c5de4ff8f0b7","abstract_canon_sha256":"f8504436be8fba71fcf7282fbeebec24f69ae1b5c30bf29783b345c6ac336563"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:27:48.137653Z","signature_b64":"3tUx4ilSNe+/7sFlu0vKNL/iXzTAWmZXo6kupOuJ9fgCYdP0WXZfNhCS1yMx8ltz85Syqj9Y/D2gqCfQDNIIBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8e183d453d8c653260b8c2a21849935547e1f54fab2f2fee0dfdbb771ce6821f","last_reissued_at":"2026-05-18T00:27:48.137160Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:27:48.137160Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Andrew Howard, Benoit Jacob, Bo Chen, Dmitry Kalenichenko, Hartwig Adam, Matthew Tang, Menglong Zhu, Skirmantas Kligys","submitted_at":"2017-12-15T23:56:52Z","abstract_excerpt":"The rising popularity of intelligent mobile devices and the daunting computational cost of deep learning-based models call for efficient and accurate on-device inference schemes. We propose a quantization scheme that allows inference to be carried out using integer-only arithmetic, which can be implemented more efficiently than floating point inference on commonly available integer-only hardware. We also co-design a training procedure to preserve end-to-end model accuracy post quantization. As a result, the proposed quantization scheme improves the tradeoff between accuracy and on-device laten"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1712.05877","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":""},"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":"1712.05877","created_at":"2026-05-18T00:27:48.137235+00:00"},{"alias_kind":"arxiv_version","alias_value":"1712.05877v1","created_at":"2026-05-18T00:27:48.137235+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1712.05877","created_at":"2026-05-18T00:27:48.137235+00:00"},{"alias_kind":"pith_short_12","alias_value":"RYMD2RJ5RRST","created_at":"2026-05-18T12:31:43.269735+00:00"},{"alias_kind":"pith_short_16","alias_value":"RYMD2RJ5RRSTEYFY","created_at":"2026-05-18T12:31:43.269735+00:00"},{"alias_kind":"pith_short_8","alias_value":"RYMD2RJ5","created_at":"2026-05-18T12:31:43.269735+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"2402.16391","citing_title":"Industry Practitioners Perspectives on AI Model Quality: Perceptions, Challenges, and Solutions","ref_index":70,"is_internal_anchor":true},{"citing_arxiv_id":"2605.20706","citing_title":"Llamas on the Web: Memory-Efficient, Performance-Portable, and Multi-Precision LLM Inference with WebGPU","ref_index":30,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10886","citing_title":"LoKA: Low-precision Kernel Applications for Recommendation Models At Scale","ref_index":40,"is_internal_anchor":true},{"citing_arxiv_id":"2208.07339","citing_title":"LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10886","citing_title":"LoKA: Low-precision Kernel Applications for Recommendation Models At Scale","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20079","citing_title":"On the Quantization Robustness of Diffusion Language Models in Coding Benchmarks","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RYMD2RJ5RRSTEYFYYKRBQSMTKV","json":"https://pith.science/pith/RYMD2RJ5RRSTEYFYYKRBQSMTKV.json","graph_json":"https://pith.science/api/pith-number/RYMD2RJ5RRSTEYFYYKRBQSMTKV/graph.json","events_json":"https://pith.science/api/pith-number/RYMD2RJ5RRSTEYFYYKRBQSMTKV/events.json","paper":"https://pith.science/paper/RYMD2RJ5"},"agent_actions":{"view_html":"https://pith.science/pith/RYMD2RJ5RRSTEYFYYKRBQSMTKV","download_json":"https://pith.science/pith/RYMD2RJ5RRSTEYFYYKRBQSMTKV.json","view_paper":"https://pith.science/paper/RYMD2RJ5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1712.05877&json=true","fetch_graph":"https://pith.science/api/pith-number/RYMD2RJ5RRSTEYFYYKRBQSMTKV/graph.json","fetch_events":"https://pith.science/api/pith-number/RYMD2RJ5RRSTEYFYYKRBQSMTKV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RYMD2RJ5RRSTEYFYYKRBQSMTKV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RYMD2RJ5RRSTEYFYYKRBQSMTKV/action/storage_attestation","attest_author":"https://pith.science/pith/RYMD2RJ5RRSTEYFYYKRBQSMTKV/action/author_attestation","sign_citation":"https://pith.science/pith/RYMD2RJ5RRSTEYFYYKRBQSMTKV/action/citation_signature","submit_replication":"https://pith.science/pith/RYMD2RJ5RRSTEYFYYKRBQSMTKV/action/replication_record"}},"created_at":"2026-05-18T00:27:48.137235+00:00","updated_at":"2026-05-18T00:27:48.137235+00:00"}