{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:74VMUVAZKFXKS7C6OJ6VMPDQYU","short_pith_number":"pith:74VMUVAZ","schema_version":"1.0","canonical_sha256":"ff2aca5419516ea97c5e727d563c70c5212ae67771ab044f91b0eb9449561bb8","source":{"kind":"arxiv","id":"1901.09504","version":3},"attestation_state":"computed","paper":{"title":"Improving Neural Network Quantization without Retraining using Outlier Channel Splitting","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Christopher De Sa, Jordan Dotzel, Ritchie Zhao, Yuwei Hu, Zhiru Zhang","submitted_at":"2019-01-28T03:50:35Z","abstract_excerpt":"Quantization can improve the execution latency and energy efficiency of neural networks on both commodity GPUs and specialized accelerators. The majority of existing literature focuses on training quantized DNNs, while this work examines the less-studied topic of quantizing a floating-point model without (re)training. DNN weights and activations follow a bell-shaped distribution post-training, while practical hardware uses a linear quantization grid. This leads to challenges in dealing with outliers in the distribution. Prior work has addressed this by clipping the outliers or using specialize"},"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":"1901.09504","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-28T03:50:35Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"d7c0bfc89beaadf5730832533a9c92ddd6dc16bd5db4a5a5959eac18d53af593","abstract_canon_sha256":"c1be50396117c145353ad491da1b9240426c5388d17a6b9ca4166b390d62a7d0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:45:20.320669Z","signature_b64":"zs0meN4XDBDOVRUqnSpMI+CfRg0MpEJ5e65bhksGik8RNtkXX9EQPDNdIp3FkPAVsgz5AuqMTR3cjJE0kpiOAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ff2aca5419516ea97c5e727d563c70c5212ae67771ab044f91b0eb9449561bb8","last_reissued_at":"2026-05-17T23:45:20.320012Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:45:20.320012Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Neural Network Quantization without Retraining using Outlier Channel Splitting","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Christopher De Sa, Jordan Dotzel, Ritchie Zhao, Yuwei Hu, Zhiru Zhang","submitted_at":"2019-01-28T03:50:35Z","abstract_excerpt":"Quantization can improve the execution latency and energy efficiency of neural networks on both commodity GPUs and specialized accelerators. The majority of existing literature focuses on training quantized DNNs, while this work examines the less-studied topic of quantizing a floating-point model without (re)training. DNN weights and activations follow a bell-shaped distribution post-training, while practical hardware uses a linear quantization grid. This leads to challenges in dealing with outliers in the distribution. Prior work has addressed this by clipping the outliers or using specialize"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1901.09504","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":""},"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":"1901.09504","created_at":"2026-05-17T23:45:20.320094+00:00"},{"alias_kind":"arxiv_version","alias_value":"1901.09504v3","created_at":"2026-05-17T23:45:20.320094+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1901.09504","created_at":"2026-05-17T23:45:20.320094+00:00"},{"alias_kind":"pith_short_12","alias_value":"74VMUVAZKFXK","created_at":"2026-05-18T12:33:12.712433+00:00"},{"alias_kind":"pith_short_16","alias_value":"74VMUVAZKFXKS7C6","created_at":"2026-05-18T12:33:12.712433+00:00"},{"alias_kind":"pith_short_8","alias_value":"74VMUVAZ","created_at":"2026-05-18T12:33:12.712433+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/74VMUVAZKFXKS7C6OJ6VMPDQYU","json":"https://pith.science/pith/74VMUVAZKFXKS7C6OJ6VMPDQYU.json","graph_json":"https://pith.science/api/pith-number/74VMUVAZKFXKS7C6OJ6VMPDQYU/graph.json","events_json":"https://pith.science/api/pith-number/74VMUVAZKFXKS7C6OJ6VMPDQYU/events.json","paper":"https://pith.science/paper/74VMUVAZ"},"agent_actions":{"view_html":"https://pith.science/pith/74VMUVAZKFXKS7C6OJ6VMPDQYU","download_json":"https://pith.science/pith/74VMUVAZKFXKS7C6OJ6VMPDQYU.json","view_paper":"https://pith.science/paper/74VMUVAZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1901.09504&json=true","fetch_graph":"https://pith.science/api/pith-number/74VMUVAZKFXKS7C6OJ6VMPDQYU/graph.json","fetch_events":"https://pith.science/api/pith-number/74VMUVAZKFXKS7C6OJ6VMPDQYU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/74VMUVAZKFXKS7C6OJ6VMPDQYU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/74VMUVAZKFXKS7C6OJ6VMPDQYU/action/storage_attestation","attest_author":"https://pith.science/pith/74VMUVAZKFXKS7C6OJ6VMPDQYU/action/author_attestation","sign_citation":"https://pith.science/pith/74VMUVAZKFXKS7C6OJ6VMPDQYU/action/citation_signature","submit_replication":"https://pith.science/pith/74VMUVAZKFXKS7C6OJ6VMPDQYU/action/replication_record"}},"created_at":"2026-05-17T23:45:20.320094+00:00","updated_at":"2026-05-17T23:45:20.320094+00:00"}