{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:USWZJMVMNRT6UNYVFYBEVYHVEE","short_pith_number":"pith:USWZJMVM","schema_version":"1.0","canonical_sha256":"a4ad94b2ac6c67ea37152e024ae0f52134ce8f24d453ddb212c13444456fde32","source":{"kind":"arxiv","id":"2405.14597","version":2},"attestation_state":"computed","paper":{"title":"Integer Scale: A Free Lunch for Faster Fine-grained Quantization of LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bo Zhang, Lin Ma, Qingyuan Li, Ran Meng, Yerui Sun, Yiduo Li, Yifan Lu, Yuchen Xie","submitted_at":"2024-05-23T14:12:58Z","abstract_excerpt":"We introduce Integer Scale, a novel post-training quantization scheme for large language models that effectively resolves the inference bottleneck in current fine-grained quantization approaches while maintaining similar accuracies. Integer Scale is a free lunch as it requires no extra calibration or fine-tuning which will otherwise incur additional costs. It can be used plug-and-play for most fine-grained quantization methods. Its integration results in at most 1.85x end-to-end speed boost over the original counterpart with comparable accuracy. Additionally, due to the orchestration of the pr"},"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":"2405.14597","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-05-23T14:12:58Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"aa2c182101a6cccdb185d49347f5e2de0aa5cfba572f88c12c07fc9b66f609d9","abstract_canon_sha256":"6eaf75cb66ed16326c84a36ae529b017c6a9562b5f6cb84ee685fdb1507f691e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:24:06.239933Z","signature_b64":"TZVX/GkwN1/l8czcObMjdNuiERqZ1HAtGWigMVxkvAvslxPP6SEDz2kPTZuV8vidxe9pXBhJ72Ab2C1T1frKDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a4ad94b2ac6c67ea37152e024ae0f52134ce8f24d453ddb212c13444456fde32","last_reissued_at":"2026-07-05T08:24:06.239440Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:24:06.239440Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Integer Scale: A Free Lunch for Faster Fine-grained Quantization of LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bo Zhang, Lin Ma, Qingyuan Li, Ran Meng, Yerui Sun, Yiduo Li, Yifan Lu, Yuchen Xie","submitted_at":"2024-05-23T14:12:58Z","abstract_excerpt":"We introduce Integer Scale, a novel post-training quantization scheme for large language models that effectively resolves the inference bottleneck in current fine-grained quantization approaches while maintaining similar accuracies. Integer Scale is a free lunch as it requires no extra calibration or fine-tuning which will otherwise incur additional costs. It can be used plug-and-play for most fine-grained quantization methods. Its integration results in at most 1.85x end-to-end speed boost over the original counterpart with comparable accuracy. Additionally, due to the orchestration of the pr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.14597","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/2405.14597/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":"2405.14597","created_at":"2026-07-05T08:24:06.239509+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.14597v2","created_at":"2026-07-05T08:24:06.239509+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.14597","created_at":"2026-07-05T08:24:06.239509+00:00"},{"alias_kind":"pith_short_12","alias_value":"USWZJMVMNRT6","created_at":"2026-07-05T08:24:06.239509+00:00"},{"alias_kind":"pith_short_16","alias_value":"USWZJMVMNRT6UNYV","created_at":"2026-07-05T08:24:06.239509+00:00"},{"alias_kind":"pith_short_8","alias_value":"USWZJMVM","created_at":"2026-07-05T08:24:06.239509+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/USWZJMVMNRT6UNYVFYBEVYHVEE","json":"https://pith.science/pith/USWZJMVMNRT6UNYVFYBEVYHVEE.json","graph_json":"https://pith.science/api/pith-number/USWZJMVMNRT6UNYVFYBEVYHVEE/graph.json","events_json":"https://pith.science/api/pith-number/USWZJMVMNRT6UNYVFYBEVYHVEE/events.json","paper":"https://pith.science/paper/USWZJMVM"},"agent_actions":{"view_html":"https://pith.science/pith/USWZJMVMNRT6UNYVFYBEVYHVEE","download_json":"https://pith.science/pith/USWZJMVMNRT6UNYVFYBEVYHVEE.json","view_paper":"https://pith.science/paper/USWZJMVM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.14597&json=true","fetch_graph":"https://pith.science/api/pith-number/USWZJMVMNRT6UNYVFYBEVYHVEE/graph.json","fetch_events":"https://pith.science/api/pith-number/USWZJMVMNRT6UNYVFYBEVYHVEE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/USWZJMVMNRT6UNYVFYBEVYHVEE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/USWZJMVMNRT6UNYVFYBEVYHVEE/action/storage_attestation","attest_author":"https://pith.science/pith/USWZJMVMNRT6UNYVFYBEVYHVEE/action/author_attestation","sign_citation":"https://pith.science/pith/USWZJMVMNRT6UNYVFYBEVYHVEE/action/citation_signature","submit_replication":"https://pith.science/pith/USWZJMVMNRT6UNYVFYBEVYHVEE/action/replication_record"}},"created_at":"2026-07-05T08:24:06.239509+00:00","updated_at":"2026-07-05T08:24:06.239509+00:00"}