{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:SL7H6JHZMZFEPY5FMWUTS4QLVD","short_pith_number":"pith:SL7H6JHZ","schema_version":"1.0","canonical_sha256":"92fe7f24f9664a47e3a565a939720ba8c3e30017679dfd30a63921caa03cf365","source":{"kind":"arxiv","id":"2307.03493","version":2},"attestation_state":"computed","paper":{"title":"ITA: An Energy-Efficient Attention and Softmax Accelerator for Quantized Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AR","authors_text":"Angelo Garofalo, Gamze \\.Islamo\\u{g}lu, Gianna Paulin, Luca Benini, Moritz Scherer, Tim Fischer, Victor J.B. Jung","submitted_at":"2023-07-07T10:05:38Z","abstract_excerpt":"Transformer networks have emerged as the state-of-the-art approach for natural language processing tasks and are gaining popularity in other domains such as computer vision and audio processing. However, the efficient hardware acceleration of transformer models poses new challenges due to their high arithmetic intensities, large memory requirements, and complex dataflow dependencies. In this work, we propose ITA, a novel accelerator architecture for transformers and related models that targets efficient inference on embedded systems by exploiting 8-bit quantization and an innovative softmax im"},"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":"2307.03493","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2023-07-07T10:05:38Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"8e3fb934ff9040ee0685345fade091ddb5e2069407bdc734b0772f6e6e143c54","abstract_canon_sha256":"416d13f3b3ecdd9a9cfd5c6cdef5261076b5839cfab3ac173356f8d5d100fced"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:48:38.820190Z","signature_b64":"X9Ezuigsyb1j7j2iBU3lUANNFtgzOIENWNU5vkIocnmkBPFJhi+oeBjxgWMLHPKQPpg4Kqe0XKNc2HleYqfyAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"92fe7f24f9664a47e3a565a939720ba8c3e30017679dfd30a63921caa03cf365","last_reissued_at":"2026-07-05T08:48:38.819813Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:48:38.819813Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ITA: An Energy-Efficient Attention and Softmax Accelerator for Quantized Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AR","authors_text":"Angelo Garofalo, Gamze \\.Islamo\\u{g}lu, Gianna Paulin, Luca Benini, Moritz Scherer, Tim Fischer, Victor J.B. Jung","submitted_at":"2023-07-07T10:05:38Z","abstract_excerpt":"Transformer networks have emerged as the state-of-the-art approach for natural language processing tasks and are gaining popularity in other domains such as computer vision and audio processing. However, the efficient hardware acceleration of transformer models poses new challenges due to their high arithmetic intensities, large memory requirements, and complex dataflow dependencies. In this work, we propose ITA, a novel accelerator architecture for transformers and related models that targets efficient inference on embedded systems by exploiting 8-bit quantization and an innovative softmax im"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2307.03493","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/2307.03493/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":"2307.03493","created_at":"2026-07-05T08:48:38.819866+00:00"},{"alias_kind":"arxiv_version","alias_value":"2307.03493v2","created_at":"2026-07-05T08:48:38.819866+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.03493","created_at":"2026-07-05T08:48:38.819866+00:00"},{"alias_kind":"pith_short_12","alias_value":"SL7H6JHZMZFE","created_at":"2026-07-05T08:48:38.819866+00:00"},{"alias_kind":"pith_short_16","alias_value":"SL7H6JHZMZFEPY5F","created_at":"2026-07-05T08:48:38.819866+00:00"},{"alias_kind":"pith_short_8","alias_value":"SL7H6JHZ","created_at":"2026-07-05T08:48:38.819866+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2504.17333","citing_title":"Fine-Grained Fusion: The Missing Piece in Area-Efficient State Space Model Acceleration","ref_index":28,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SL7H6JHZMZFEPY5FMWUTS4QLVD","json":"https://pith.science/pith/SL7H6JHZMZFEPY5FMWUTS4QLVD.json","graph_json":"https://pith.science/api/pith-number/SL7H6JHZMZFEPY5FMWUTS4QLVD/graph.json","events_json":"https://pith.science/api/pith-number/SL7H6JHZMZFEPY5FMWUTS4QLVD/events.json","paper":"https://pith.science/paper/SL7H6JHZ"},"agent_actions":{"view_html":"https://pith.science/pith/SL7H6JHZMZFEPY5FMWUTS4QLVD","download_json":"https://pith.science/pith/SL7H6JHZMZFEPY5FMWUTS4QLVD.json","view_paper":"https://pith.science/paper/SL7H6JHZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2307.03493&json=true","fetch_graph":"https://pith.science/api/pith-number/SL7H6JHZMZFEPY5FMWUTS4QLVD/graph.json","fetch_events":"https://pith.science/api/pith-number/SL7H6JHZMZFEPY5FMWUTS4QLVD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SL7H6JHZMZFEPY5FMWUTS4QLVD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SL7H6JHZMZFEPY5FMWUTS4QLVD/action/storage_attestation","attest_author":"https://pith.science/pith/SL7H6JHZMZFEPY5FMWUTS4QLVD/action/author_attestation","sign_citation":"https://pith.science/pith/SL7H6JHZMZFEPY5FMWUTS4QLVD/action/citation_signature","submit_replication":"https://pith.science/pith/SL7H6JHZMZFEPY5FMWUTS4QLVD/action/replication_record"}},"created_at":"2026-07-05T08:48:38.819866+00:00","updated_at":"2026-07-05T08:48:38.819866+00:00"}