{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:HDTEO7UZHLSAJZ6FKYY2Y5ES2P","short_pith_number":"pith:HDTEO7UZ","schema_version":"1.0","canonical_sha256":"38e6477e993ae404e7c55631ac7492d3fdf52efde2a451ad452586145cd4eb13","source":{"kind":"arxiv","id":"2506.03781","version":2},"attestation_state":"computed","paper":{"title":"Unifying Uniform and Binary-coding Quantization for Accurate Compression of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Beomseok Kwon, Byeongwook Kim, Dongsoo Lee, Jeongin Bae, Minjun Kim, Se Jung Kwon, Seungcheol Park, U Kang","submitted_at":"2025-06-04T09:42:17Z","abstract_excerpt":"How can we quantize large language models while preserving accuracy? Quantization is essential for deploying large language models (LLMs) efficiently. Binary-coding quantization (BCQ) and uniform quantization (UQ) are promising quantization schemes that have strong expressiveness and optimizability, respectively. However, neither scheme leverages both advantages. In this paper, we propose UniQuanF (Unified Quantization with Flexible Mapping), an accurate quantization method for LLMs. UniQuanF harnesses both strong expressiveness and optimizability by unifying the flexible mapping technique in "},"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":"2506.03781","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-04T09:42:17Z","cross_cats_sorted":[],"title_canon_sha256":"1eb48181c788fedebcbf3736d338e1ea2f024c63076873d7e15db4200025972b","abstract_canon_sha256":"9cf835f04b88a354ff6d514c4ec789cefe52472c0d05bafaf3a14abb06fed53a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:22:18.864677Z","signature_b64":"JSucsF0AE5na0IEB8GXfzBy111jMlvnAe1audq5rnHAx8EB0mvcy7W3FscSTMBY9j1WNNHbpE4KiQ19aHQBhDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"38e6477e993ae404e7c55631ac7492d3fdf52efde2a451ad452586145cd4eb13","last_reissued_at":"2026-07-05T11:22:18.864167Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:22:18.864167Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unifying Uniform and Binary-coding Quantization for Accurate Compression of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Beomseok Kwon, Byeongwook Kim, Dongsoo Lee, Jeongin Bae, Minjun Kim, Se Jung Kwon, Seungcheol Park, U Kang","submitted_at":"2025-06-04T09:42:17Z","abstract_excerpt":"How can we quantize large language models while preserving accuracy? Quantization is essential for deploying large language models (LLMs) efficiently. Binary-coding quantization (BCQ) and uniform quantization (UQ) are promising quantization schemes that have strong expressiveness and optimizability, respectively. However, neither scheme leverages both advantages. In this paper, we propose UniQuanF (Unified Quantization with Flexible Mapping), an accurate quantization method for LLMs. UniQuanF harnesses both strong expressiveness and optimizability by unifying the flexible mapping technique in "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.03781","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/2506.03781/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":"2506.03781","created_at":"2026-07-05T11:22:18.864226+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.03781v2","created_at":"2026-07-05T11:22:18.864226+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.03781","created_at":"2026-07-05T11:22:18.864226+00:00"},{"alias_kind":"pith_short_12","alias_value":"HDTEO7UZHLSA","created_at":"2026-07-05T11:22:18.864226+00:00"},{"alias_kind":"pith_short_16","alias_value":"HDTEO7UZHLSAJZ6F","created_at":"2026-07-05T11:22:18.864226+00:00"},{"alias_kind":"pith_short_8","alias_value":"HDTEO7UZ","created_at":"2026-07-05T11:22:18.864226+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.04050","citing_title":"LiftQuant: Continuous Bit-Width LLM via Dimensional Lifting and Projection","ref_index":58,"is_internal_anchor":false},{"citing_arxiv_id":"2606.04050","citing_title":"LiftQuant: Continuous Bit-Width LLM via Dimensional Lifting and Projection","ref_index":58,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19167","citing_title":"LBLLM: Lightweight Binarization of Large Language Models via Three-Stage Distillation","ref_index":99,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HDTEO7UZHLSAJZ6FKYY2Y5ES2P","json":"https://pith.science/pith/HDTEO7UZHLSAJZ6FKYY2Y5ES2P.json","graph_json":"https://pith.science/api/pith-number/HDTEO7UZHLSAJZ6FKYY2Y5ES2P/graph.json","events_json":"https://pith.science/api/pith-number/HDTEO7UZHLSAJZ6FKYY2Y5ES2P/events.json","paper":"https://pith.science/paper/HDTEO7UZ"},"agent_actions":{"view_html":"https://pith.science/pith/HDTEO7UZHLSAJZ6FKYY2Y5ES2P","download_json":"https://pith.science/pith/HDTEO7UZHLSAJZ6FKYY2Y5ES2P.json","view_paper":"https://pith.science/paper/HDTEO7UZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.03781&json=true","fetch_graph":"https://pith.science/api/pith-number/HDTEO7UZHLSAJZ6FKYY2Y5ES2P/graph.json","fetch_events":"https://pith.science/api/pith-number/HDTEO7UZHLSAJZ6FKYY2Y5ES2P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HDTEO7UZHLSAJZ6FKYY2Y5ES2P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HDTEO7UZHLSAJZ6FKYY2Y5ES2P/action/storage_attestation","attest_author":"https://pith.science/pith/HDTEO7UZHLSAJZ6FKYY2Y5ES2P/action/author_attestation","sign_citation":"https://pith.science/pith/HDTEO7UZHLSAJZ6FKYY2Y5ES2P/action/citation_signature","submit_replication":"https://pith.science/pith/HDTEO7UZHLSAJZ6FKYY2Y5ES2P/action/replication_record"}},"created_at":"2026-07-05T11:22:18.864226+00:00","updated_at":"2026-07-05T11:22:18.864226+00:00"}