{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:EBWWTWDE7PG3SCB2K7KUNHLJOI","short_pith_number":"pith:EBWWTWDE","schema_version":"1.0","canonical_sha256":"206d69d864fbcdb9083a57d5469d69723f3adadfea0c9096917cf62c8cb5bd37","source":{"kind":"arxiv","id":"2506.11784","version":1},"attestation_state":"computed","paper":{"title":"GPLQ: A General, Practical, and Lightning QAT Method for Vision Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Guang Liang, Jianxin Wu, Xinyao Liu","submitted_at":"2025-06-13T13:45:17Z","abstract_excerpt":"Vision Transformers (ViTs) are essential in computer vision but are computationally intensive, too. Model quantization, particularly to low bit-widths like 4-bit, aims to alleviate this difficulty, yet existing Post-Training Quantization (PTQ) and Quantization-Aware Training (QAT) methods exhibit significant limitations. PTQ often incurs substantial accuracy drop, while QAT achieves high accuracy but suffers from prohibitive computational costs, limited generalization to downstream tasks, training instability, and lacking of open-source codebase. To address these challenges, this paper introdu"},"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":"2506.11784","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-06-13T13:45:17Z","cross_cats_sorted":[],"title_canon_sha256":"ef19f58e4389952811c6c6983228f59ec876066f084c00658b17e32dd67ceaa5","abstract_canon_sha256":"18507c01ae2b65715e4a81bb84616b3fa53581d133410094cb2460b5d16d162c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:21:10.677101Z","signature_b64":"k19eB+6Xg5o8YNrK1abnq+yDix/GQQq/MP6Vrq11VvGM2He/M407iL5UY0nQQf63SdDPYdo8CkX6TBCgCSzXDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"206d69d864fbcdb9083a57d5469d69723f3adadfea0c9096917cf62c8cb5bd37","last_reissued_at":"2026-07-05T11:21:10.676692Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:21:10.676692Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GPLQ: A General, Practical, and Lightning QAT Method for Vision Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Guang Liang, Jianxin Wu, Xinyao Liu","submitted_at":"2025-06-13T13:45:17Z","abstract_excerpt":"Vision Transformers (ViTs) are essential in computer vision but are computationally intensive, too. Model quantization, particularly to low bit-widths like 4-bit, aims to alleviate this difficulty, yet existing Post-Training Quantization (PTQ) and Quantization-Aware Training (QAT) methods exhibit significant limitations. PTQ often incurs substantial accuracy drop, while QAT achieves high accuracy but suffers from prohibitive computational costs, limited generalization to downstream tasks, training instability, and lacking of open-source codebase. To address these challenges, this paper introdu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.11784","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/2506.11784/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":"2506.11784","created_at":"2026-07-05T11:21:10.676750+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.11784v1","created_at":"2026-07-05T11:21:10.676750+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.11784","created_at":"2026-07-05T11:21:10.676750+00:00"},{"alias_kind":"pith_short_12","alias_value":"EBWWTWDE7PG3","created_at":"2026-07-05T11:21:10.676750+00:00"},{"alias_kind":"pith_short_16","alias_value":"EBWWTWDE7PG3SCB2","created_at":"2026-07-05T11:21:10.676750+00:00"},{"alias_kind":"pith_short_8","alias_value":"EBWWTWDE","created_at":"2026-07-05T11:21:10.676750+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.16423","citing_title":"Nonlinear Bipolar Compensation: Handling Outliers in Post-Training Quantization","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01330","citing_title":"Colinearity Decay: Training Quantization-Friendly ViTs with Outlier Decay","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EBWWTWDE7PG3SCB2K7KUNHLJOI","json":"https://pith.science/pith/EBWWTWDE7PG3SCB2K7KUNHLJOI.json","graph_json":"https://pith.science/api/pith-number/EBWWTWDE7PG3SCB2K7KUNHLJOI/graph.json","events_json":"https://pith.science/api/pith-number/EBWWTWDE7PG3SCB2K7KUNHLJOI/events.json","paper":"https://pith.science/paper/EBWWTWDE"},"agent_actions":{"view_html":"https://pith.science/pith/EBWWTWDE7PG3SCB2K7KUNHLJOI","download_json":"https://pith.science/pith/EBWWTWDE7PG3SCB2K7KUNHLJOI.json","view_paper":"https://pith.science/paper/EBWWTWDE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.11784&json=true","fetch_graph":"https://pith.science/api/pith-number/EBWWTWDE7PG3SCB2K7KUNHLJOI/graph.json","fetch_events":"https://pith.science/api/pith-number/EBWWTWDE7PG3SCB2K7KUNHLJOI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EBWWTWDE7PG3SCB2K7KUNHLJOI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EBWWTWDE7PG3SCB2K7KUNHLJOI/action/storage_attestation","attest_author":"https://pith.science/pith/EBWWTWDE7PG3SCB2K7KUNHLJOI/action/author_attestation","sign_citation":"https://pith.science/pith/EBWWTWDE7PG3SCB2K7KUNHLJOI/action/citation_signature","submit_replication":"https://pith.science/pith/EBWWTWDE7PG3SCB2K7KUNHLJOI/action/replication_record"}},"created_at":"2026-07-05T11:21:10.676750+00:00","updated_at":"2026-07-05T11:21:10.676750+00:00"}