{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:ACTTRRPGFCK5XYWWSEEYVZQDVB","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"26f56a9b3d1a42d7e562e1925063a867ab87233c446fc13aadb58db069748d68","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-01-27T22:44:18Z","title_canon_sha256":"b3371183b37756d1a652b5918654abda4079e05b302ba2ae4dfbdfa0d137d36c"},"schema_version":"1.0","source":{"id":"2301.12017","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2301.12017","created_at":"2026-07-05T06:15:45Z"},{"alias_kind":"arxiv_version","alias_value":"2301.12017v2","created_at":"2026-07-05T06:15:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.12017","created_at":"2026-07-05T06:15:45Z"},{"alias_kind":"pith_short_12","alias_value":"ACTTRRPGFCK5","created_at":"2026-07-05T06:15:45Z"},{"alias_kind":"pith_short_16","alias_value":"ACTTRRPGFCK5XYWW","created_at":"2026-07-05T06:15:45Z"},{"alias_kind":"pith_short_8","alias_value":"ACTTRRPG","created_at":"2026-07-05T06:15:45Z"}],"graph_snapshots":[{"event_id":"sha256:4864a3aa7ccf0bbb75197af7a1e41bc65206f4f910da7666d2c9940638731276","target":"graph","created_at":"2026-07-05T06:15:45Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2301.12017/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Improving the deployment efficiency of transformer-based language models has been challenging given their high computation and memory cost. While INT8 quantization has recently been shown to be effective in reducing both the memory cost and latency while preserving model accuracy, it remains unclear whether we can leverage INT4 (which doubles peak hardware throughput) to achieve further latency improvement. In this study, we explore the feasibility of employing INT4 weight and activation (W4A4) quantization for language models. Our findings indicate that W4A4 quantization introduces no to negl","authors_text":"Cheng Li, Reza Yazdani Aminabadi, Xiaoxia Wu, Yuxiong He, Zhewei Yao","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-01-27T22:44:18Z","title":"Understanding INT4 Quantization for Transformer Models: Latency Speedup, Composability, and Failure Cases"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.12017","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:b70443000f666384d9ae1bcf454e7d3f80203a4cf9fbb97dcaa203f76d26bcec","target":"record","created_at":"2026-07-05T06:15:45Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"26f56a9b3d1a42d7e562e1925063a867ab87233c446fc13aadb58db069748d68","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-01-27T22:44:18Z","title_canon_sha256":"b3371183b37756d1a652b5918654abda4079e05b302ba2ae4dfbdfa0d137d36c"},"schema_version":"1.0","source":{"id":"2301.12017","kind":"arxiv","version":2}},"canonical_sha256":"00a738c5e62895dbe2d691098ae603a8577e6d26927dbf449083fd281beccb22","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"00a738c5e62895dbe2d691098ae603a8577e6d26927dbf449083fd281beccb22","first_computed_at":"2026-07-05T06:15:45.176039Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:15:45.176039Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Aye2azMrQTkqqm02MkJCvr8AxYdA3nj65w8DIMkdKFm6pilsE8dL/UeML3lWG4T0yH1f6nnOENGFIPIxnYDxBw==","signature_status":"signed_v1","signed_at":"2026-07-05T06:15:45.176565Z","signed_message":"canonical_sha256_bytes"},"source_id":"2301.12017","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b70443000f666384d9ae1bcf454e7d3f80203a4cf9fbb97dcaa203f76d26bcec","sha256:4864a3aa7ccf0bbb75197af7a1e41bc65206f4f910da7666d2c9940638731276"],"state_sha256":"ea563801aa50b3514e0c79e3f4bf70ce153ff17faf42f8aebc3eea8818abab76"}