{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YPLGILLRBAAZLLXKLX2QEKZC6K","short_pith_number":"pith:YPLGILLR","schema_version":"1.0","canonical_sha256":"c3d6642d71080195aeea5df5022b22f287e8d4cc1f4ead89d2c6e96d32a6120c","source":{"kind":"arxiv","id":"2505.05530","version":1},"attestation_state":"computed","paper":{"title":"Low-bit Model Quantization for Deep Neural Networks: A Survey","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Guihai Chen, Haotong Qin, Kai Liu, Kaiwen Tao, Linghe Kong, Qian Zheng, Wenbo Li, Xianglong Liu, Xiaokang Yang, Yong Guo, Yulun Zhang, Zhiteng Li","submitted_at":"2025-05-08T13:26:19Z","abstract_excerpt":"With unprecedented rapid development, deep neural networks (DNNs) have deeply influenced almost all fields. However, their heavy computation costs and model sizes are usually unacceptable in real-world deployment. Model quantization, an effective weight-lighting technique, has become an indispensable procedure in the whole deployment pipeline. The essence of quantization acceleration is the conversion from continuous floating-point numbers to discrete integer ones, which significantly speeds up the memory I/O and calculation, i.e., addition and multiplication. However, performance degradation "},"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":"2505.05530","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-08T13:26:19Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"78ff81763ba50ad6ea8859feb2bcd7b11961d7d5501a6dec71ae47d812008c38","abstract_canon_sha256":"3d8f764017dc964b326ebed85194cce1b382ebfba5d85ffdcd03d895805e0450"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:00:27.469537Z","signature_b64":"7PKYNOvMcNIFtinfMhaiVnCC206/Mf/DPKTFIMTeNrhVQNoUIS6BFtnm4xf3f1/bmDpjrKI74GIfFZyjee3EDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c3d6642d71080195aeea5df5022b22f287e8d4cc1f4ead89d2c6e96d32a6120c","last_reissued_at":"2026-07-05T11:00:27.469014Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:00:27.469014Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Low-bit Model Quantization for Deep Neural Networks: A Survey","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Guihai Chen, Haotong Qin, Kai Liu, Kaiwen Tao, Linghe Kong, Qian Zheng, Wenbo Li, Xianglong Liu, Xiaokang Yang, Yong Guo, Yulun Zhang, Zhiteng Li","submitted_at":"2025-05-08T13:26:19Z","abstract_excerpt":"With unprecedented rapid development, deep neural networks (DNNs) have deeply influenced almost all fields. However, their heavy computation costs and model sizes are usually unacceptable in real-world deployment. Model quantization, an effective weight-lighting technique, has become an indispensable procedure in the whole deployment pipeline. The essence of quantization acceleration is the conversion from continuous floating-point numbers to discrete integer ones, which significantly speeds up the memory I/O and calculation, i.e., addition and multiplication. However, performance degradation "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.05530","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/2505.05530/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":"2505.05530","created_at":"2026-07-05T11:00:27.469075+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.05530v1","created_at":"2026-07-05T11:00:27.469075+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.05530","created_at":"2026-07-05T11:00:27.469075+00:00"},{"alias_kind":"pith_short_12","alias_value":"YPLGILLRBAAZ","created_at":"2026-07-05T11:00:27.469075+00:00"},{"alias_kind":"pith_short_16","alias_value":"YPLGILLRBAAZLLXK","created_at":"2026-07-05T11:00:27.469075+00:00"},{"alias_kind":"pith_short_8","alias_value":"YPLGILLR","created_at":"2026-07-05T11:00:27.469075+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.28421","citing_title":"JuZhou 1.0 Technical Report: The First Edge-Native Text-to-Image Foundation Model Trained Entirely on China-Developed AI Accelerators","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10933","citing_title":"DECO: Sparse Mixture-of-Experts with Dense-Comparable Performance on End-Side Devices","ref_index":179,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17997","citing_title":"MARR: Module-Adaptive Residual Reconstruction for Low-Bit Post-Training Quantization","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2604.03420","citing_title":"Zero-Shot Quantization via Weight-Space Arithmetic","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10933","citing_title":"DECO: Sparse Mixture-of-Experts with Dense-Comparable Performance on End-Side Devices","ref_index":179,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10933","citing_title":"DECO: Sparse Mixture-of-Experts with Dense-Comparable Performance on End-Side Devices","ref_index":179,"is_internal_anchor":false},{"citing_arxiv_id":"2604.09220","citing_title":"TinyNeRV: Compact Neural Video Representations via Capacity Scaling, Distillation, and Low-Precision Inference","ref_index":60,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YPLGILLRBAAZLLXKLX2QEKZC6K","json":"https://pith.science/pith/YPLGILLRBAAZLLXKLX2QEKZC6K.json","graph_json":"https://pith.science/api/pith-number/YPLGILLRBAAZLLXKLX2QEKZC6K/graph.json","events_json":"https://pith.science/api/pith-number/YPLGILLRBAAZLLXKLX2QEKZC6K/events.json","paper":"https://pith.science/paper/YPLGILLR"},"agent_actions":{"view_html":"https://pith.science/pith/YPLGILLRBAAZLLXKLX2QEKZC6K","download_json":"https://pith.science/pith/YPLGILLRBAAZLLXKLX2QEKZC6K.json","view_paper":"https://pith.science/paper/YPLGILLR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.05530&json=true","fetch_graph":"https://pith.science/api/pith-number/YPLGILLRBAAZLLXKLX2QEKZC6K/graph.json","fetch_events":"https://pith.science/api/pith-number/YPLGILLRBAAZLLXKLX2QEKZC6K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YPLGILLRBAAZLLXKLX2QEKZC6K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YPLGILLRBAAZLLXKLX2QEKZC6K/action/storage_attestation","attest_author":"https://pith.science/pith/YPLGILLRBAAZLLXKLX2QEKZC6K/action/author_attestation","sign_citation":"https://pith.science/pith/YPLGILLRBAAZLLXKLX2QEKZC6K/action/citation_signature","submit_replication":"https://pith.science/pith/YPLGILLRBAAZLLXKLX2QEKZC6K/action/replication_record"}},"created_at":"2026-07-05T11:00:27.469075+00:00","updated_at":"2026-07-05T11:00:27.469075+00:00"}