{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:C6C2HQX6M4Z2YCHROHSBZUCPXB","short_pith_number":"pith:C6C2HQX6","schema_version":"1.0","canonical_sha256":"1785a3c2fe6733ac08f171e41cd04fb8510af7b8f29c06bcdcf2120db02a491b","source":{"kind":"arxiv","id":"2403.06408","version":1},"attestation_state":"computed","paper":{"title":"What Makes Quantization for Large Language Models Hard? An Empirical Study from the Lens of Perturbation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Dongyan Zhao, Jiahao Liu, Jingang Wang, Rui Yan, Xunliang Cai, Zhuocheng Gong","submitted_at":"2024-03-11T03:42:51Z","abstract_excerpt":"Quantization has emerged as a promising technique for improving the memory and computational efficiency of large language models (LLMs). Though the trade-off between performance and efficiency is well-known, there is still much to be learned about the relationship between quantization and LLM performance. To shed light on this relationship, we propose a new perspective on quantization, viewing it as perturbations added to the weights and activations of LLMs. We call this approach \"the lens of perturbation\". Using this lens, we conduct experiments with various artificial perturbations to explor"},"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":"2403.06408","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-11T03:42:51Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a6f7bb63650e43a390c3b7047e20442526d7a09f4c1abee7c5963b3ffff3e779","abstract_canon_sha256":"b47b3bcf8f26e65e5231b25d955547010c5799ca7f97ae8dd47047481d1c3158"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:54:37.576445Z","signature_b64":"PXuhffuBqAU3Tkst7TOBnW9x02iLDWpaBNxhCvfrDKw3QrGY3fhjXoyGXV5ZJt2ecDpl6kwrfZ1h43/bNSF7Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1785a3c2fe6733ac08f171e41cd04fb8510af7b8f29c06bcdcf2120db02a491b","last_reissued_at":"2026-07-05T07:54:37.575940Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:54:37.575940Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"What Makes Quantization for Large Language Models Hard? An Empirical Study from the Lens of Perturbation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Dongyan Zhao, Jiahao Liu, Jingang Wang, Rui Yan, Xunliang Cai, Zhuocheng Gong","submitted_at":"2024-03-11T03:42:51Z","abstract_excerpt":"Quantization has emerged as a promising technique for improving the memory and computational efficiency of large language models (LLMs). Though the trade-off between performance and efficiency is well-known, there is still much to be learned about the relationship between quantization and LLM performance. To shed light on this relationship, we propose a new perspective on quantization, viewing it as perturbations added to the weights and activations of LLMs. We call this approach \"the lens of perturbation\". Using this lens, we conduct experiments with various artificial perturbations to explor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.06408","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/2403.06408/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":"2403.06408","created_at":"2026-07-05T07:54:37.575997+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.06408v1","created_at":"2026-07-05T07:54:37.575997+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.06408","created_at":"2026-07-05T07:54:37.575997+00:00"},{"alias_kind":"pith_short_12","alias_value":"C6C2HQX6M4Z2","created_at":"2026-07-05T07:54:37.575997+00:00"},{"alias_kind":"pith_short_16","alias_value":"C6C2HQX6M4Z2YCHR","created_at":"2026-07-05T07:54:37.575997+00:00"},{"alias_kind":"pith_short_8","alias_value":"C6C2HQX6","created_at":"2026-07-05T07:54:37.575997+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2409.00084","citing_title":"Vision-Language and Large Language Model Performance in Gastroenterology: GPT, Claude, Llama, Phi, Mistral, Gemma, and Quantized Models","ref_index":32,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/C6C2HQX6M4Z2YCHROHSBZUCPXB","json":"https://pith.science/pith/C6C2HQX6M4Z2YCHROHSBZUCPXB.json","graph_json":"https://pith.science/api/pith-number/C6C2HQX6M4Z2YCHROHSBZUCPXB/graph.json","events_json":"https://pith.science/api/pith-number/C6C2HQX6M4Z2YCHROHSBZUCPXB/events.json","paper":"https://pith.science/paper/C6C2HQX6"},"agent_actions":{"view_html":"https://pith.science/pith/C6C2HQX6M4Z2YCHROHSBZUCPXB","download_json":"https://pith.science/pith/C6C2HQX6M4Z2YCHROHSBZUCPXB.json","view_paper":"https://pith.science/paper/C6C2HQX6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.06408&json=true","fetch_graph":"https://pith.science/api/pith-number/C6C2HQX6M4Z2YCHROHSBZUCPXB/graph.json","fetch_events":"https://pith.science/api/pith-number/C6C2HQX6M4Z2YCHROHSBZUCPXB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/C6C2HQX6M4Z2YCHROHSBZUCPXB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/C6C2HQX6M4Z2YCHROHSBZUCPXB/action/storage_attestation","attest_author":"https://pith.science/pith/C6C2HQX6M4Z2YCHROHSBZUCPXB/action/author_attestation","sign_citation":"https://pith.science/pith/C6C2HQX6M4Z2YCHROHSBZUCPXB/action/citation_signature","submit_replication":"https://pith.science/pith/C6C2HQX6M4Z2YCHROHSBZUCPXB/action/replication_record"}},"created_at":"2026-07-05T07:54:37.575997+00:00","updated_at":"2026-07-05T07:54:37.575997+00:00"}