{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:T5CUYY4Y3B45CLVO4BD7AHERXZ","short_pith_number":"pith:T5CUYY4Y","schema_version":"1.0","canonical_sha256":"9f454c6398d879d12eaee047f01c91be48d390b81b01457b1572cc06d97b51e6","source":{"kind":"arxiv","id":"2311.13290","version":2},"attestation_state":"computed","paper":{"title":"Hyft: A Reconfigurable Softmax Accelerator with Hybrid Numeric Format for both Training and Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Sai Qian Zhang, Tianhua Xia","submitted_at":"2023-11-22T10:19:40Z","abstract_excerpt":"The attention mechanism is a pivotal element within the transformer architecture, making a substantial contribution to its exceptional performance. Within this attention mechanism, Softmax is an imperative component that enables the model to assess the degree of correlation between various segments of the input. Yet, prior research has shown that Softmax operations can significantly increase processing latency and energy consumption in the transformer network due to their internal nonlinear operations and data dependencies. In this work, we proposed Hyft, a hardware efficient floating point So"},"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":"2311.13290","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2023-11-22T10:19:40Z","cross_cats_sorted":[],"title_canon_sha256":"034cd63118e143683e07c2474cd9d60d1baabc8744243fa5747ec18d5ccc1026","abstract_canon_sha256":"0fe91c2db9f299f46cdd82edb82195a84080b24cf518955c743bfecd3128dc16"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:02:55.311330Z","signature_b64":"7x7bjKkGXi6XcgesJLlY7voZFcRGSMzwybbpOsRuNH3b88kXP5xc6sMg0YR5JJnj71Ofy2AOKQn1sS8c0+JcCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9f454c6398d879d12eaee047f01c91be48d390b81b01457b1572cc06d97b51e6","last_reissued_at":"2026-07-05T09:02:55.310803Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:02:55.310803Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Hyft: A Reconfigurable Softmax Accelerator with Hybrid Numeric Format for both Training and Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Sai Qian Zhang, Tianhua Xia","submitted_at":"2023-11-22T10:19:40Z","abstract_excerpt":"The attention mechanism is a pivotal element within the transformer architecture, making a substantial contribution to its exceptional performance. Within this attention mechanism, Softmax is an imperative component that enables the model to assess the degree of correlation between various segments of the input. Yet, prior research has shown that Softmax operations can significantly increase processing latency and energy consumption in the transformer network due to their internal nonlinear operations and data dependencies. In this work, we proposed Hyft, a hardware efficient floating point So"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.13290","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/2311.13290/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":"2311.13290","created_at":"2026-07-05T09:02:55.310873+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.13290v2","created_at":"2026-07-05T09:02:55.310873+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.13290","created_at":"2026-07-05T09:02:55.310873+00:00"},{"alias_kind":"pith_short_12","alias_value":"T5CUYY4Y3B45","created_at":"2026-07-05T09:02:55.310873+00:00"},{"alias_kind":"pith_short_16","alias_value":"T5CUYY4Y3B45CLVO","created_at":"2026-07-05T09:02:55.310873+00:00"},{"alias_kind":"pith_short_8","alias_value":"T5CUYY4Y","created_at":"2026-07-05T09:02:55.310873+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.00026","citing_title":"Pushing the Limits of BFP on Narrow Precision LLM Inference","ref_index":34,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/T5CUYY4Y3B45CLVO4BD7AHERXZ","json":"https://pith.science/pith/T5CUYY4Y3B45CLVO4BD7AHERXZ.json","graph_json":"https://pith.science/api/pith-number/T5CUYY4Y3B45CLVO4BD7AHERXZ/graph.json","events_json":"https://pith.science/api/pith-number/T5CUYY4Y3B45CLVO4BD7AHERXZ/events.json","paper":"https://pith.science/paper/T5CUYY4Y"},"agent_actions":{"view_html":"https://pith.science/pith/T5CUYY4Y3B45CLVO4BD7AHERXZ","download_json":"https://pith.science/pith/T5CUYY4Y3B45CLVO4BD7AHERXZ.json","view_paper":"https://pith.science/paper/T5CUYY4Y","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.13290&json=true","fetch_graph":"https://pith.science/api/pith-number/T5CUYY4Y3B45CLVO4BD7AHERXZ/graph.json","fetch_events":"https://pith.science/api/pith-number/T5CUYY4Y3B45CLVO4BD7AHERXZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/T5CUYY4Y3B45CLVO4BD7AHERXZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/T5CUYY4Y3B45CLVO4BD7AHERXZ/action/storage_attestation","attest_author":"https://pith.science/pith/T5CUYY4Y3B45CLVO4BD7AHERXZ/action/author_attestation","sign_citation":"https://pith.science/pith/T5CUYY4Y3B45CLVO4BD7AHERXZ/action/citation_signature","submit_replication":"https://pith.science/pith/T5CUYY4Y3B45CLVO4BD7AHERXZ/action/replication_record"}},"created_at":"2026-07-05T09:02:55.310873+00:00","updated_at":"2026-07-05T09:02:55.310873+00:00"}