{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:73YGDMMPOOTYCZGLND6YX3U6BO","short_pith_number":"pith:73YGDMMP","schema_version":"1.0","canonical_sha256":"fef061b18f73a78164cb68fd8bee9e0ba2b1e272b827cd8d80a10f7c0023f312","source":{"kind":"arxiv","id":"1910.06188","version":2},"attestation_state":"computed","paper":{"title":"Q8BERT: Quantized 8Bit BERT","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Guy Boudoukh, Moshe Wasserblat, Ofir Zafrir, Peter Izsak","submitted_at":"2019-10-14T14:55:19Z","abstract_excerpt":"Recently, pre-trained Transformer based language models such as BERT and GPT, have shown great improvement in many Natural Language Processing (NLP) tasks. However, these models contain a large amount of parameters. The emergence of even larger and more accurate models such as GPT2 and Megatron, suggest a trend of large pre-trained Transformer models. However, using these large models in production environments is a complex task requiring a large amount of compute, memory and power resources. In this work we show how to perform quantization-aware training during the fine-tuning phase of BERT i"},"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":"1910.06188","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-10-14T14:55:19Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"a845d8ba2ec038e865dcde6bd5ecc38c2bc4641c01be634b3cbd0d1e1418f17d","abstract_canon_sha256":"f08dedee89e4a054181e1e34d7b4cc1d48c1a241d745301d4998a8906106ba1c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:41:36.700303Z","signature_b64":"6W4L9ivoQ8Fqaeq8cHEBidwpZ32uq6+ZXnX5pB5l+XanXv0+6ZjhTqOxUIhQ39f7TUSJkQNjGXhWO1Ax0mjnCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fef061b18f73a78164cb68fd8bee9e0ba2b1e272b827cd8d80a10f7c0023f312","last_reissued_at":"2026-07-05T03:41:36.699945Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:41:36.699945Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Q8BERT: Quantized 8Bit BERT","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Guy Boudoukh, Moshe Wasserblat, Ofir Zafrir, Peter Izsak","submitted_at":"2019-10-14T14:55:19Z","abstract_excerpt":"Recently, pre-trained Transformer based language models such as BERT and GPT, have shown great improvement in many Natural Language Processing (NLP) tasks. However, these models contain a large amount of parameters. The emergence of even larger and more accurate models such as GPT2 and Megatron, suggest a trend of large pre-trained Transformer models. However, using these large models in production environments is a complex task requiring a large amount of compute, memory and power resources. In this work we show how to perform quantization-aware training during the fine-tuning phase of BERT i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1910.06188","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/1910.06188/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":"1910.06188","created_at":"2026-07-05T03:41:36.700004+00:00"},{"alias_kind":"arxiv_version","alias_value":"1910.06188v2","created_at":"2026-07-05T03:41:36.700004+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1910.06188","created_at":"2026-07-05T03:41:36.700004+00:00"},{"alias_kind":"pith_short_12","alias_value":"73YGDMMPOOTY","created_at":"2026-07-05T03:41:36.700004+00:00"},{"alias_kind":"pith_short_16","alias_value":"73YGDMMPOOTYCZGL","created_at":"2026-07-05T03:41:36.700004+00:00"},{"alias_kind":"pith_short_8","alias_value":"73YGDMMP","created_at":"2026-07-05T03:41:36.700004+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/73YGDMMPOOTYCZGLND6YX3U6BO","json":"https://pith.science/pith/73YGDMMPOOTYCZGLND6YX3U6BO.json","graph_json":"https://pith.science/api/pith-number/73YGDMMPOOTYCZGLND6YX3U6BO/graph.json","events_json":"https://pith.science/api/pith-number/73YGDMMPOOTYCZGLND6YX3U6BO/events.json","paper":"https://pith.science/paper/73YGDMMP"},"agent_actions":{"view_html":"https://pith.science/pith/73YGDMMPOOTYCZGLND6YX3U6BO","download_json":"https://pith.science/pith/73YGDMMPOOTYCZGLND6YX3U6BO.json","view_paper":"https://pith.science/paper/73YGDMMP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1910.06188&json=true","fetch_graph":"https://pith.science/api/pith-number/73YGDMMPOOTYCZGLND6YX3U6BO/graph.json","fetch_events":"https://pith.science/api/pith-number/73YGDMMPOOTYCZGLND6YX3U6BO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/73YGDMMPOOTYCZGLND6YX3U6BO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/73YGDMMPOOTYCZGLND6YX3U6BO/action/storage_attestation","attest_author":"https://pith.science/pith/73YGDMMPOOTYCZGLND6YX3U6BO/action/author_attestation","sign_citation":"https://pith.science/pith/73YGDMMPOOTYCZGLND6YX3U6BO/action/citation_signature","submit_replication":"https://pith.science/pith/73YGDMMPOOTYCZGLND6YX3U6BO/action/replication_record"}},"created_at":"2026-07-05T03:41:36.700004+00:00","updated_at":"2026-07-05T03:41:36.700004+00:00"}