{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:IBNDVO7ISZSUDK7BCJ2GLFKMHJ","short_pith_number":"pith:IBNDVO7I","schema_version":"1.0","canonical_sha256":"405a3abbe8966541abe1127465954c3a640aa7978355a33a0397dda5e9adc94c","source":{"kind":"arxiv","id":"2605.18475","version":1},"attestation_state":"computed","paper":{"title":"GAMMA: Global Bit Allocation for Mixed-Precision Models under Arbitrary Budgets","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Haiyan Zhao, Haoyu Wang, Lihua Zhang, Tianbo Huang, Xu Han, Zhangyang Yao","submitted_at":"2026-05-18T14:30:58Z","abstract_excerpt":"Mixed-precision quantization improves the budget--accuracy trade-off for large language models (LLMs) by allocating more bits to sensitive modules. However, automating this allocation at LLM scale faces a unique combination of constraints: learnable approaches require quantization-aware training, which is infeasible for billion-parameter models; training-free alternatives rely on static proxy metrics that miss cross-module interactions and must be recomputed per target budget; and search-based methods are expensive without guaranteeing exact budget compliance. We propose GAMMA, a quantizer-agn"},"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":"2605.18475","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-18T14:30:58Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a473c0f7f22c129057f58389f2158d8d52dec4ca417813b2e676c19d15998ff3","abstract_canon_sha256":"24496d9d1fe7ba40e06e04fc2e50821cb0f7d166edd232017cfdc6592f7d5526"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:06:03.121698Z","signature_b64":"csgQj7bZ0H8Ry6co4WexzCRE3uddXgNizFfvJu76ttoXoGbyvnowL94qBdrKdyjS6F+lUNYJ5DaW4sneZT/aDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"405a3abbe8966541abe1127465954c3a640aa7978355a33a0397dda5e9adc94c","last_reissued_at":"2026-05-20T00:06:03.120928Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:06:03.120928Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GAMMA: Global Bit Allocation for Mixed-Precision Models under Arbitrary Budgets","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Haiyan Zhao, Haoyu Wang, Lihua Zhang, Tianbo Huang, Xu Han, Zhangyang Yao","submitted_at":"2026-05-18T14:30:58Z","abstract_excerpt":"Mixed-precision quantization improves the budget--accuracy trade-off for large language models (LLMs) by allocating more bits to sensitive modules. However, automating this allocation at LLM scale faces a unique combination of constraints: learnable approaches require quantization-aware training, which is infeasible for billion-parameter models; training-free alternatives rely on static proxy metrics that miss cross-module interactions and must be recomputed per target budget; and search-based methods are expensive without guaranteeing exact budget compliance. We propose GAMMA, a quantizer-agn"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.18475","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/2605.18475/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":"2605.18475","created_at":"2026-05-20T00:06:03.121064+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.18475v1","created_at":"2026-05-20T00:06:03.121064+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.18475","created_at":"2026-05-20T00:06:03.121064+00:00"},{"alias_kind":"pith_short_12","alias_value":"IBNDVO7ISZSU","created_at":"2026-05-20T00:06:03.121064+00:00"},{"alias_kind":"pith_short_16","alias_value":"IBNDVO7ISZSUDK7B","created_at":"2026-05-20T00:06:03.121064+00:00"},{"alias_kind":"pith_short_8","alias_value":"IBNDVO7I","created_at":"2026-05-20T00:06:03.121064+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/IBNDVO7ISZSUDK7BCJ2GLFKMHJ","json":"https://pith.science/pith/IBNDVO7ISZSUDK7BCJ2GLFKMHJ.json","graph_json":"https://pith.science/api/pith-number/IBNDVO7ISZSUDK7BCJ2GLFKMHJ/graph.json","events_json":"https://pith.science/api/pith-number/IBNDVO7ISZSUDK7BCJ2GLFKMHJ/events.json","paper":"https://pith.science/paper/IBNDVO7I"},"agent_actions":{"view_html":"https://pith.science/pith/IBNDVO7ISZSUDK7BCJ2GLFKMHJ","download_json":"https://pith.science/pith/IBNDVO7ISZSUDK7BCJ2GLFKMHJ.json","view_paper":"https://pith.science/paper/IBNDVO7I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.18475&json=true","fetch_graph":"https://pith.science/api/pith-number/IBNDVO7ISZSUDK7BCJ2GLFKMHJ/graph.json","fetch_events":"https://pith.science/api/pith-number/IBNDVO7ISZSUDK7BCJ2GLFKMHJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IBNDVO7ISZSUDK7BCJ2GLFKMHJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IBNDVO7ISZSUDK7BCJ2GLFKMHJ/action/storage_attestation","attest_author":"https://pith.science/pith/IBNDVO7ISZSUDK7BCJ2GLFKMHJ/action/author_attestation","sign_citation":"https://pith.science/pith/IBNDVO7ISZSUDK7BCJ2GLFKMHJ/action/citation_signature","submit_replication":"https://pith.science/pith/IBNDVO7ISZSUDK7BCJ2GLFKMHJ/action/replication_record"}},"created_at":"2026-05-20T00:06:03.121064+00:00","updated_at":"2026-05-20T00:06:03.121064+00:00"}