{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:LNNWZB3QYRWKOQNADU432ZMBP7","short_pith_number":"pith:LNNWZB3Q","schema_version":"1.0","canonical_sha256":"5b5b6c8770c46ca741a01d39bd65817fe19ba73df00e5093ab4dee594ec986c6","source":{"kind":"arxiv","id":"2409.17092","version":2},"attestation_state":"computed","paper":{"title":"Accumulator-Aware Post-Training Quantization for Large Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.DM"],"primary_cat":"cs.LG","authors_text":"Fabian Grob, Giuseppe Franco, Ian Colbert, Jinjie Zhang, Rayan Saab","submitted_at":"2024-09-25T16:58:35Z","abstract_excerpt":"When quantizing weights and activations to increasingly narrower representations, the cost of additions begins to dominate that of multiplications in multiply-accumulate (MAC) units. Recent studies show that reducing addition costs via low-precision accumulation improves throughput, power, and area across inference platforms, albeit with an increased risk of overflow. Accumulator-aware quantization research has so far only considered the quantization-aware training (QAT) paradigm, in which models are fine-tuned or trained from scratch with quantization in the loop. As models and datasets conti"},"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":"2409.17092","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-25T16:58:35Z","cross_cats_sorted":["cs.AI","cs.DM"],"title_canon_sha256":"434f793e21ebb99b17f57b45924dc135d68fda132fe734e63a599accc76bfb44","abstract_canon_sha256":"826c29910ffca51d00bafebf7f98f98f8435222be76a918fe449ff15ad7e48b3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:45:53.777416Z","signature_b64":"sqsQNoY7NpBE+J2GOdzm2lbLVvL7SLMi4axzM1EiGDLgDsKqfqK3X8zf8QX3CFz5yoGkky0SswiJ0jWItZZMBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5b5b6c8770c46ca741a01d39bd65817fe19ba73df00e5093ab4dee594ec986c6","last_reissued_at":"2026-07-05T11:45:53.776887Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:45:53.776887Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Accumulator-Aware Post-Training Quantization for Large Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.DM"],"primary_cat":"cs.LG","authors_text":"Fabian Grob, Giuseppe Franco, Ian Colbert, Jinjie Zhang, Rayan Saab","submitted_at":"2024-09-25T16:58:35Z","abstract_excerpt":"When quantizing weights and activations to increasingly narrower representations, the cost of additions begins to dominate that of multiplications in multiply-accumulate (MAC) units. Recent studies show that reducing addition costs via low-precision accumulation improves throughput, power, and area across inference platforms, albeit with an increased risk of overflow. Accumulator-aware quantization research has so far only considered the quantization-aware training (QAT) paradigm, in which models are fine-tuned or trained from scratch with quantization in the loop. As models and datasets conti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.17092","kind":"arxiv","version":2},"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/2409.17092/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":"2409.17092","created_at":"2026-07-05T11:45:53.776949+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.17092v2","created_at":"2026-07-05T11:45:53.776949+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.17092","created_at":"2026-07-05T11:45:53.776949+00:00"},{"alias_kind":"pith_short_12","alias_value":"LNNWZB3QYRWK","created_at":"2026-07-05T11:45:53.776949+00:00"},{"alias_kind":"pith_short_16","alias_value":"LNNWZB3QYRWKOQNA","created_at":"2026-07-05T11:45:53.776949+00:00"},{"alias_kind":"pith_short_8","alias_value":"LNNWZB3Q","created_at":"2026-07-05T11:45:53.776949+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.18184","citing_title":"Unified Stochastic Framework for Neural Network Quantization and Pruning","ref_index":6,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LNNWZB3QYRWKOQNADU432ZMBP7","json":"https://pith.science/pith/LNNWZB3QYRWKOQNADU432ZMBP7.json","graph_json":"https://pith.science/api/pith-number/LNNWZB3QYRWKOQNADU432ZMBP7/graph.json","events_json":"https://pith.science/api/pith-number/LNNWZB3QYRWKOQNADU432ZMBP7/events.json","paper":"https://pith.science/paper/LNNWZB3Q"},"agent_actions":{"view_html":"https://pith.science/pith/LNNWZB3QYRWKOQNADU432ZMBP7","download_json":"https://pith.science/pith/LNNWZB3QYRWKOQNADU432ZMBP7.json","view_paper":"https://pith.science/paper/LNNWZB3Q","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.17092&json=true","fetch_graph":"https://pith.science/api/pith-number/LNNWZB3QYRWKOQNADU432ZMBP7/graph.json","fetch_events":"https://pith.science/api/pith-number/LNNWZB3QYRWKOQNADU432ZMBP7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LNNWZB3QYRWKOQNADU432ZMBP7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LNNWZB3QYRWKOQNADU432ZMBP7/action/storage_attestation","attest_author":"https://pith.science/pith/LNNWZB3QYRWKOQNADU432ZMBP7/action/author_attestation","sign_citation":"https://pith.science/pith/LNNWZB3QYRWKOQNADU432ZMBP7/action/citation_signature","submit_replication":"https://pith.science/pith/LNNWZB3QYRWKOQNADU432ZMBP7/action/replication_record"}},"created_at":"2026-07-05T11:45:53.776949+00:00","updated_at":"2026-07-05T11:45:53.776949+00:00"}