{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:JSPRXWWQ2DWQXQ4RAFXY4QEEXA","short_pith_number":"pith:JSPRXWWQ","schema_version":"1.0","canonical_sha256":"4c9f1bdad0d0ed0bc391016f8e4084b8125a937291d2df6e38f5440333a90e31","source":{"kind":"arxiv","id":"2410.13056","version":4},"attestation_state":"computed","paper":{"title":"Channel-Wise Mixed-Precision Quantization for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bike Xie, Cong Shen, Jundong Li, Zihan Chen","submitted_at":"2024-10-16T21:34:41Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated remarkable success across a wide range of language tasks, but their deployment on edge devices remains challenging due to the substantial memory requirements imposed by their large parameter sizes. Weight-only quantization presents a promising solution to reduce the memory footprint of LLMs. However, existing approaches primarily focus on integer-bit quantization, limiting their adaptability to fractional-bit quantization tasks and preventing the full utilization of available storage space on devices. In this paper, we introduce Channel-Wise Mixed"},"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":"2410.13056","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-16T21:34:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"8a055e424e4319669d158c69a51b4765f201918fc07c770dd0968182e7784417","abstract_canon_sha256":"cf69e9793b5d78b55ded78eb83311b97a4e2a8a38e0e570e1f9918dd9473dc25"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-05T01:14:24.641113Z","signature_b64":"4738V90Zq74gPWgJDzAB4AzbLolsrqtxHvtcb+/M2OM5TuefMg52wtlNTq70oX4hwIBRdErqdHNCAuqfWr8/CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4c9f1bdad0d0ed0bc391016f8e4084b8125a937291d2df6e38f5440333a90e31","last_reissued_at":"2026-06-05T01:14:24.640309Z","signature_status":"signed_v1","first_computed_at":"2026-06-05T01:14:24.640309Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Channel-Wise Mixed-Precision Quantization for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bike Xie, Cong Shen, Jundong Li, Zihan Chen","submitted_at":"2024-10-16T21:34:41Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated remarkable success across a wide range of language tasks, but their deployment on edge devices remains challenging due to the substantial memory requirements imposed by their large parameter sizes. Weight-only quantization presents a promising solution to reduce the memory footprint of LLMs. However, existing approaches primarily focus on integer-bit quantization, limiting their adaptability to fractional-bit quantization tasks and preventing the full utilization of available storage space on devices. In this paper, we introduce Channel-Wise Mixed"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.13056","kind":"arxiv","version":4},"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/2410.13056/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":"2410.13056","created_at":"2026-06-05T01:14:24.640421+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.13056v4","created_at":"2026-06-05T01:14:24.640421+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.13056","created_at":"2026-06-05T01:14:24.640421+00:00"},{"alias_kind":"pith_short_12","alias_value":"JSPRXWWQ2DWQ","created_at":"2026-06-05T01:14:24.640421+00:00"},{"alias_kind":"pith_short_16","alias_value":"JSPRXWWQ2DWQXQ4R","created_at":"2026-06-05T01:14:24.640421+00:00"},{"alias_kind":"pith_short_8","alias_value":"JSPRXWWQ","created_at":"2026-06-05T01:14:24.640421+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"2605.24011","citing_title":"ActQuant: Sub-4-bit Action-Guided Quantization for Vision-Language-Action Models","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2605.25054","citing_title":"Scale When Needed: Adaptive Neuron-level Mixed Precision Quantization Aware Training","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"2604.09083","citing_title":"EdgeFlow: Fast Cold Starts for LLMs on Mobile Devices","ref_index":14,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JSPRXWWQ2DWQXQ4RAFXY4QEEXA","json":"https://pith.science/pith/JSPRXWWQ2DWQXQ4RAFXY4QEEXA.json","graph_json":"https://pith.science/api/pith-number/JSPRXWWQ2DWQXQ4RAFXY4QEEXA/graph.json","events_json":"https://pith.science/api/pith-number/JSPRXWWQ2DWQXQ4RAFXY4QEEXA/events.json","paper":"https://pith.science/paper/JSPRXWWQ"},"agent_actions":{"view_html":"https://pith.science/pith/JSPRXWWQ2DWQXQ4RAFXY4QEEXA","download_json":"https://pith.science/pith/JSPRXWWQ2DWQXQ4RAFXY4QEEXA.json","view_paper":"https://pith.science/paper/JSPRXWWQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.13056&json=true","fetch_graph":"https://pith.science/api/pith-number/JSPRXWWQ2DWQXQ4RAFXY4QEEXA/graph.json","fetch_events":"https://pith.science/api/pith-number/JSPRXWWQ2DWQXQ4RAFXY4QEEXA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JSPRXWWQ2DWQXQ4RAFXY4QEEXA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JSPRXWWQ2DWQXQ4RAFXY4QEEXA/action/storage_attestation","attest_author":"https://pith.science/pith/JSPRXWWQ2DWQXQ4RAFXY4QEEXA/action/author_attestation","sign_citation":"https://pith.science/pith/JSPRXWWQ2DWQXQ4RAFXY4QEEXA/action/citation_signature","submit_replication":"https://pith.science/pith/JSPRXWWQ2DWQXQ4RAFXY4QEEXA/action/replication_record"}},"created_at":"2026-06-05T01:14:24.640421+00:00","updated_at":"2026-06-05T01:14:24.640421+00:00"}