{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:XNLZRIZT4L7UZLIJUXOATU7LPW","short_pith_number":"pith:XNLZRIZT","schema_version":"1.0","canonical_sha256":"bb5798a333e2ff4cad09a5dc09d3eb7d9c088ad1853583fe63211943c5744d74","source":{"kind":"arxiv","id":"2605.29756","version":1},"attestation_state":"computed","paper":{"title":"LFQ: Logit-aware Final-block Quantization for Boosting the Generation Quality of Low-Bit Quantized LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Eunho Yang, June Yong Yang, Jung Hyun Lee, Jungwook Choi","submitted_at":"2026-05-28T11:02:23Z","abstract_excerpt":"As large language models continue to scale, low-bit weight-only post-training quantization (PTQ) offers a practical solution to their memory-efficient deployment. Although block-wise PTQ is capable of matching the full-precision (FP) baseline on basic language modeling and understanding, its quality is degraded for generative tasks -- especially at longer responses and extended chains of thought, which is critical in boosting task accuracy. We attribute this shortfall to two factors: (i) the omission of the unembedding layer (the LM head) in block-wise optimization and (ii) the reliance on the"},"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":"2605.29756","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-28T11:02:23Z","cross_cats_sorted":[],"title_canon_sha256":"73efe6e53d8e9413883fe8809ae7fe793dfb5348391d2e30ad45801810e66195","abstract_canon_sha256":"539ba42ed6458577f866401f1e393d7b9ecd746a8e39e51bf6a952c324678b3d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:05:58.055596Z","signature_b64":"5cyyCpoaV8I/1X7F0dPX9dxYr2bJHDODnyp0uyho3eML0Se87Im/HZQH8kKpPwtcyX5R3nMymSXpNNp8r4wdAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bb5798a333e2ff4cad09a5dc09d3eb7d9c088ad1853583fe63211943c5744d74","last_reissued_at":"2026-05-29T01:05:58.054350Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:05:58.054350Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LFQ: Logit-aware Final-block Quantization for Boosting the Generation Quality of Low-Bit Quantized LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Eunho Yang, June Yong Yang, Jung Hyun Lee, Jungwook Choi","submitted_at":"2026-05-28T11:02:23Z","abstract_excerpt":"As large language models continue to scale, low-bit weight-only post-training quantization (PTQ) offers a practical solution to their memory-efficient deployment. Although block-wise PTQ is capable of matching the full-precision (FP) baseline on basic language modeling and understanding, its quality is degraded for generative tasks -- especially at longer responses and extended chains of thought, which is critical in boosting task accuracy. We attribute this shortfall to two factors: (i) the omission of the unembedding layer (the LM head) in block-wise optimization and (ii) the reliance on the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.29756","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/2605.29756/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":"2605.29756","created_at":"2026-05-29T01:05:58.055134+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.29756v1","created_at":"2026-05-29T01:05:58.055134+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.29756","created_at":"2026-05-29T01:05:58.055134+00:00"},{"alias_kind":"pith_short_12","alias_value":"XNLZRIZT4L7U","created_at":"2026-05-29T01:05:58.055134+00:00"},{"alias_kind":"pith_short_16","alias_value":"XNLZRIZT4L7UZLIJ","created_at":"2026-05-29T01:05:58.055134+00:00"},{"alias_kind":"pith_short_8","alias_value":"XNLZRIZT","created_at":"2026-05-29T01:05:58.055134+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XNLZRIZT4L7UZLIJUXOATU7LPW","json":"https://pith.science/pith/XNLZRIZT4L7UZLIJUXOATU7LPW.json","graph_json":"https://pith.science/api/pith-number/XNLZRIZT4L7UZLIJUXOATU7LPW/graph.json","events_json":"https://pith.science/api/pith-number/XNLZRIZT4L7UZLIJUXOATU7LPW/events.json","paper":"https://pith.science/paper/XNLZRIZT"},"agent_actions":{"view_html":"https://pith.science/pith/XNLZRIZT4L7UZLIJUXOATU7LPW","download_json":"https://pith.science/pith/XNLZRIZT4L7UZLIJUXOATU7LPW.json","view_paper":"https://pith.science/paper/XNLZRIZT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.29756&json=true","fetch_graph":"https://pith.science/api/pith-number/XNLZRIZT4L7UZLIJUXOATU7LPW/graph.json","fetch_events":"https://pith.science/api/pith-number/XNLZRIZT4L7UZLIJUXOATU7LPW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XNLZRIZT4L7UZLIJUXOATU7LPW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XNLZRIZT4L7UZLIJUXOATU7LPW/action/storage_attestation","attest_author":"https://pith.science/pith/XNLZRIZT4L7UZLIJUXOATU7LPW/action/author_attestation","sign_citation":"https://pith.science/pith/XNLZRIZT4L7UZLIJUXOATU7LPW/action/citation_signature","submit_replication":"https://pith.science/pith/XNLZRIZT4L7UZLIJUXOATU7LPW/action/replication_record"}},"created_at":"2026-05-29T01:05:58.055134+00:00","updated_at":"2026-05-29T01:05:58.055134+00:00"}