{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:ATUAXOPXSBIMLSQMDEDFHHR2QG","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"a796ec1e822c3aac8046c4eaff3b14e75d03ef94877bad1f1a738a4ff3a0c539","cross_cats_sorted":["cs.AI","cs.CL","cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-11-06T17:29:24Z","title_canon_sha256":"1cb3825bb60f8bac31bd685e24bca3bb91e525c5d13739c660bce4a8991fc3ba"},"schema_version":"1.0","source":{"id":"2311.03426","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.03426","created_at":"2026-07-05T11:08:55Z"},{"alias_kind":"arxiv_version","alias_value":"2311.03426v2","created_at":"2026-07-05T11:08:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.03426","created_at":"2026-07-05T11:08:55Z"},{"alias_kind":"pith_short_12","alias_value":"ATUAXOPXSBIM","created_at":"2026-07-05T11:08:55Z"},{"alias_kind":"pith_short_16","alias_value":"ATUAXOPXSBIMLSQM","created_at":"2026-07-05T11:08:55Z"},{"alias_kind":"pith_short_8","alias_value":"ATUAXOPX","created_at":"2026-07-05T11:08:55Z"}],"graph_snapshots":[{"event_id":"sha256:a73a3f4889538acae6e3173776709bdadbbcda25829c5a3de120cefc46dad870","target":"graph","created_at":"2026-07-05T11:08:55Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2311.03426/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Massive transformer-based models face several challenges, including slow and computationally intensive pre-training and over-parametrization. This paper addresses these challenges by proposing a versatile method called GQKVA, which generalizes query, key, and value grouping techniques. GQKVA is designed to speed up transformer pre-training while reducing the model size. Our experiments with various GQKVA variants highlight a clear trade-off between performance and model size, allowing for customized choices based on resource and time limitations. Our findings also indicate that the conventiona","authors_text":"Austin Wen, Farnoosh Javadi, Foozhan Ataiefard, Habib Hajimolahoseini, Kangling Liu, Mohammad Hassanpour, Omar Mohamed Awad, Saina Asani, Walid Ahmed, Yang Liu","cross_cats":["cs.AI","cs.CL","cs.CV"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-11-06T17:29:24Z","title":"GQKVA: Efficient Pre-training of Transformers by Grouping Queries, Keys, and Values"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.03426","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:7f4f37b7cea636dce216ffb55d8aa9bc5599637bff44cdd855757550133ecf5d","target":"record","created_at":"2026-07-05T11:08:55Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"a796ec1e822c3aac8046c4eaff3b14e75d03ef94877bad1f1a738a4ff3a0c539","cross_cats_sorted":["cs.AI","cs.CL","cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-11-06T17:29:24Z","title_canon_sha256":"1cb3825bb60f8bac31bd685e24bca3bb91e525c5d13739c660bce4a8991fc3ba"},"schema_version":"1.0","source":{"id":"2311.03426","kind":"arxiv","version":2}},"canonical_sha256":"04e80bb9f79050c5ca0c1906539e3a8184d0f75d278bcff9ed55f6b034011d3f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"04e80bb9f79050c5ca0c1906539e3a8184d0f75d278bcff9ed55f6b034011d3f","first_computed_at":"2026-07-05T11:08:55.527549Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:08:55.527549Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"F1oIR3ZKSZLOh9JZx46/mb0HcqrmL3/O/WfVCERQp7L1D6enfB7uoKU2oCSw7MeOUGlyx55rHKefapC/Hsb3AQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:08:55.528050Z","signed_message":"canonical_sha256_bytes"},"source_id":"2311.03426","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7f4f37b7cea636dce216ffb55d8aa9bc5599637bff44cdd855757550133ecf5d","sha256:a73a3f4889538acae6e3173776709bdadbbcda25829c5a3de120cefc46dad870"],"state_sha256":"e1082d693cd908be8600ac66e1de32737b242c7c562396b6e180cffda3024c48"}