{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:TSTDXU6CGO66BSCZDEW6WTVJWU","short_pith_number":"pith:TSTDXU6C","schema_version":"1.0","canonical_sha256":"9ca63bd3c233bde0c859192deb4ea9b51f5258a9bb9b7c06f1c383bbe34f5d51","source":{"kind":"arxiv","id":"2509.15206","version":3},"attestation_state":"computed","paper":{"title":"Fair-GPTQ: Bias-Aware Quantization for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Guillaume Metzler, Irina Proskurina, Julien Velcin","submitted_at":"2025-09-18T17:56:16Z","abstract_excerpt":"The high memory demands of generative language models have drawn attention to quantization, which reduces memory usage by mapping model weights to lower-precision integers. However, recent empirical studies show that, while efficient, quantization can increase the likelihood of generating biased outputs and degrade performance on fairness benchmarks. In this work, we draw new links between quantization and model fairness by adding explicit group-fairness constraints to the quantization objective and introduce Fair-GPTQ, the first quantization method explicitly designed to reduce unfairness 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":"2509.15206","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-18T17:56:16Z","cross_cats_sorted":[],"title_canon_sha256":"4221419c838b001ccd69f1572d5c62eb43e939eb2ebbb54afd4b9af1a45808f4","abstract_canon_sha256":"4f49ad7b231d40d07c8b3a044c1d055cbe8a57dd4294fd5fb0b68d3086047c9b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T02:19:40.755266Z","signature_b64":"GG6xcMmA4kLPSSYE/NvzYYj2iHxABnXobfYOckPpe0XiZVHNo/8ytEeSORo4ioQcTDkM3fJmXW6YkLJzYtGkAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9ca63bd3c233bde0c859192deb4ea9b51f5258a9bb9b7c06f1c383bbe34f5d51","last_reissued_at":"2026-07-07T02:19:40.754535Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T02:19:40.754535Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fair-GPTQ: Bias-Aware Quantization for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Guillaume Metzler, Irina Proskurina, Julien Velcin","submitted_at":"2025-09-18T17:56:16Z","abstract_excerpt":"The high memory demands of generative language models have drawn attention to quantization, which reduces memory usage by mapping model weights to lower-precision integers. However, recent empirical studies show that, while efficient, quantization can increase the likelihood of generating biased outputs and degrade performance on fairness benchmarks. In this work, we draw new links between quantization and model fairness by adding explicit group-fairness constraints to the quantization objective and introduce Fair-GPTQ, the first quantization method explicitly designed to reduce unfairness in "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.15206","kind":"arxiv","version":3},"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/2509.15206/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":"2509.15206","created_at":"2026-07-07T02:19:40.754634+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.15206v3","created_at":"2026-07-07T02:19:40.754634+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.15206","created_at":"2026-07-07T02:19:40.754634+00:00"},{"alias_kind":"pith_short_12","alias_value":"TSTDXU6CGO66","created_at":"2026-07-07T02:19:40.754634+00:00"},{"alias_kind":"pith_short_16","alias_value":"TSTDXU6CGO66BSCZ","created_at":"2026-07-07T02:19:40.754634+00:00"},{"alias_kind":"pith_short_8","alias_value":"TSTDXU6C","created_at":"2026-07-07T02:19:40.754634+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/TSTDXU6CGO66BSCZDEW6WTVJWU","json":"https://pith.science/pith/TSTDXU6CGO66BSCZDEW6WTVJWU.json","graph_json":"https://pith.science/api/pith-number/TSTDXU6CGO66BSCZDEW6WTVJWU/graph.json","events_json":"https://pith.science/api/pith-number/TSTDXU6CGO66BSCZDEW6WTVJWU/events.json","paper":"https://pith.science/paper/TSTDXU6C"},"agent_actions":{"view_html":"https://pith.science/pith/TSTDXU6CGO66BSCZDEW6WTVJWU","download_json":"https://pith.science/pith/TSTDXU6CGO66BSCZDEW6WTVJWU.json","view_paper":"https://pith.science/paper/TSTDXU6C","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.15206&json=true","fetch_graph":"https://pith.science/api/pith-number/TSTDXU6CGO66BSCZDEW6WTVJWU/graph.json","fetch_events":"https://pith.science/api/pith-number/TSTDXU6CGO66BSCZDEW6WTVJWU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TSTDXU6CGO66BSCZDEW6WTVJWU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TSTDXU6CGO66BSCZDEW6WTVJWU/action/storage_attestation","attest_author":"https://pith.science/pith/TSTDXU6CGO66BSCZDEW6WTVJWU/action/author_attestation","sign_citation":"https://pith.science/pith/TSTDXU6CGO66BSCZDEW6WTVJWU/action/citation_signature","submit_replication":"https://pith.science/pith/TSTDXU6CGO66BSCZDEW6WTVJWU/action/replication_record"}},"created_at":"2026-07-07T02:19:40.754634+00:00","updated_at":"2026-07-07T02:19:40.754634+00:00"}