{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:4XB4DUTJDRU4FW3AEO3V7CVGKC","short_pith_number":"pith:4XB4DUTJ","schema_version":"1.0","canonical_sha256":"e5c3c1d2691c69c2db6023b75f8aa6508159853f29fc5328190876fe197eabb6","source":{"kind":"arxiv","id":"2411.06084","version":1},"attestation_state":"computed","paper":{"title":"Optimizing Large Language Models through Quantization: A Comparative Analysis of PTQ and QAT Techniques","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Jahid Hasan","submitted_at":"2024-11-09T06:30:13Z","abstract_excerpt":"This paper presents a comprehensive analysis of quantization techniques for optimizing Large Language Models (LLMs), specifically focusing on Post-Training Quantization (PTQ) and Quantization-Aware Training (QAT). Through empirical evaluation across models ranging from 10M to 1B parameters, we demonstrate that quantization can achieve up to 68% reduction in model size while maintaining performance within 6% of full-precision baselines when utilizing our proposed scaling factor {\\gamma}. Our experiments show that INT8 quantization delivers a 40% reduction in computational cost and power consump"},"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":"2411.06084","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-09T06:30:13Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"ed0a0d268319089542ce74bd28bbe543f33c1c5bab2fdb121d108f2715842847","abstract_canon_sha256":"918e23d0649224d31a90f1c2a2a46aa6b59573a894242e237f8eb8b7aa1a2139"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:33:16.897767Z","signature_b64":"HFsiS1e73nwnjXGOrX9jxwFSb6m1Ow8dwA3F9EjVXH64KHt6xZS7NRgQFlEbGGVgQuDzTHIiSGPad4CTvMKRBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e5c3c1d2691c69c2db6023b75f8aa6508159853f29fc5328190876fe197eabb6","last_reissued_at":"2026-07-05T09:33:16.897241Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:33:16.897241Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Optimizing Large Language Models through Quantization: A Comparative Analysis of PTQ and QAT Techniques","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Jahid Hasan","submitted_at":"2024-11-09T06:30:13Z","abstract_excerpt":"This paper presents a comprehensive analysis of quantization techniques for optimizing Large Language Models (LLMs), specifically focusing on Post-Training Quantization (PTQ) and Quantization-Aware Training (QAT). Through empirical evaluation across models ranging from 10M to 1B parameters, we demonstrate that quantization can achieve up to 68% reduction in model size while maintaining performance within 6% of full-precision baselines when utilizing our proposed scaling factor {\\gamma}. Our experiments show that INT8 quantization delivers a 40% reduction in computational cost and power consump"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.06084","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2411.06084/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":"2411.06084","created_at":"2026-07-05T09:33:16.897304+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.06084v1","created_at":"2026-07-05T09:33:16.897304+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.06084","created_at":"2026-07-05T09:33:16.897304+00:00"},{"alias_kind":"pith_short_12","alias_value":"4XB4DUTJDRU4","created_at":"2026-07-05T09:33:16.897304+00:00"},{"alias_kind":"pith_short_16","alias_value":"4XB4DUTJDRU4FW3A","created_at":"2026-07-05T09:33:16.897304+00:00"},{"alias_kind":"pith_short_8","alias_value":"4XB4DUTJ","created_at":"2026-07-05T09:33:16.897304+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.00494","citing_title":"ProjQ: Project-and-Quantize for Adapter-Aware LLM Compression","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19195","citing_title":"The Thermodynamic Costs of Simple Linear Regression","ref_index":82,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4XB4DUTJDRU4FW3AEO3V7CVGKC","json":"https://pith.science/pith/4XB4DUTJDRU4FW3AEO3V7CVGKC.json","graph_json":"https://pith.science/api/pith-number/4XB4DUTJDRU4FW3AEO3V7CVGKC/graph.json","events_json":"https://pith.science/api/pith-number/4XB4DUTJDRU4FW3AEO3V7CVGKC/events.json","paper":"https://pith.science/paper/4XB4DUTJ"},"agent_actions":{"view_html":"https://pith.science/pith/4XB4DUTJDRU4FW3AEO3V7CVGKC","download_json":"https://pith.science/pith/4XB4DUTJDRU4FW3AEO3V7CVGKC.json","view_paper":"https://pith.science/paper/4XB4DUTJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.06084&json=true","fetch_graph":"https://pith.science/api/pith-number/4XB4DUTJDRU4FW3AEO3V7CVGKC/graph.json","fetch_events":"https://pith.science/api/pith-number/4XB4DUTJDRU4FW3AEO3V7CVGKC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4XB4DUTJDRU4FW3AEO3V7CVGKC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4XB4DUTJDRU4FW3AEO3V7CVGKC/action/storage_attestation","attest_author":"https://pith.science/pith/4XB4DUTJDRU4FW3AEO3V7CVGKC/action/author_attestation","sign_citation":"https://pith.science/pith/4XB4DUTJDRU4FW3AEO3V7CVGKC/action/citation_signature","submit_replication":"https://pith.science/pith/4XB4DUTJDRU4FW3AEO3V7CVGKC/action/replication_record"}},"created_at":"2026-07-05T09:33:16.897304+00:00","updated_at":"2026-07-05T09:33:16.897304+00:00"}