{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:TWWCXSCQY3RVJLUQX3DTPLKTDO","short_pith_number":"pith:TWWCXSCQ","schema_version":"1.0","canonical_sha256":"9dac2bc850c6e354ae90bec737ad531ba8e39366b20e08df60386259eb7322ba","source":{"kind":"arxiv","id":"2401.17544","version":1},"attestation_state":"computed","paper":{"title":"Trainable Fixed-Point Quantization for Deep Learning Acceleration on FPGAs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Dingyi Dai, Jiahao Zhang, Qi Sun, Yaohui Cai, Yichi Zhang, Zhanqiu Hu, Zhiru Zhang","submitted_at":"2024-01-31T02:18:27Z","abstract_excerpt":"Quantization is a crucial technique for deploying deep learning models on resource-constrained devices, such as embedded FPGAs. Prior efforts mostly focus on quantizing matrix multiplications, leaving other layers like BatchNorm or shortcuts in floating-point form, even though fixed-point arithmetic is more efficient on FPGAs. A common practice is to fine-tune a pre-trained model to fixed-point for FPGA deployment, but potentially degrading accuracy.\n  This work presents QFX, a novel trainable fixed-point quantization approach that automatically learns the binary-point position during model tr"},"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":"2401.17544","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-01-31T02:18:27Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"1f21b5100ba7acdb67a91e8e72d00d9a6d9568c3268bfd6b36acca303b6a0b88","abstract_canon_sha256":"20cfc77384352b82310da969b94510485725056e50ffeb107ebdecff8c928bb7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:39:38.429156Z","signature_b64":"mzTUbSY7llh69P+oIMo5STP2iBvBjDbt1hL6jEjSAO4vMqEQ7hTXjL2cOezqMCRDwzI1d2s8BECeEUOjqkcEAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9dac2bc850c6e354ae90bec737ad531ba8e39366b20e08df60386259eb7322ba","last_reissued_at":"2026-07-05T07:39:38.428721Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:39:38.428721Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Trainable Fixed-Point Quantization for Deep Learning Acceleration on FPGAs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Dingyi Dai, Jiahao Zhang, Qi Sun, Yaohui Cai, Yichi Zhang, Zhanqiu Hu, Zhiru Zhang","submitted_at":"2024-01-31T02:18:27Z","abstract_excerpt":"Quantization is a crucial technique for deploying deep learning models on resource-constrained devices, such as embedded FPGAs. Prior efforts mostly focus on quantizing matrix multiplications, leaving other layers like BatchNorm or shortcuts in floating-point form, even though fixed-point arithmetic is more efficient on FPGAs. A common practice is to fine-tune a pre-trained model to fixed-point for FPGA deployment, but potentially degrading accuracy.\n  This work presents QFX, a novel trainable fixed-point quantization approach that automatically learns the binary-point position during model tr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.17544","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/2401.17544/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":"2401.17544","created_at":"2026-07-05T07:39:38.428785+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.17544v1","created_at":"2026-07-05T07:39:38.428785+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.17544","created_at":"2026-07-05T07:39:38.428785+00:00"},{"alias_kind":"pith_short_12","alias_value":"TWWCXSCQY3RV","created_at":"2026-07-05T07:39:38.428785+00:00"},{"alias_kind":"pith_short_16","alias_value":"TWWCXSCQY3RVJLUQ","created_at":"2026-07-05T07:39:38.428785+00:00"},{"alias_kind":"pith_short_8","alias_value":"TWWCXSCQ","created_at":"2026-07-05T07:39:38.428785+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/TWWCXSCQY3RVJLUQX3DTPLKTDO","json":"https://pith.science/pith/TWWCXSCQY3RVJLUQX3DTPLKTDO.json","graph_json":"https://pith.science/api/pith-number/TWWCXSCQY3RVJLUQX3DTPLKTDO/graph.json","events_json":"https://pith.science/api/pith-number/TWWCXSCQY3RVJLUQX3DTPLKTDO/events.json","paper":"https://pith.science/paper/TWWCXSCQ"},"agent_actions":{"view_html":"https://pith.science/pith/TWWCXSCQY3RVJLUQX3DTPLKTDO","download_json":"https://pith.science/pith/TWWCXSCQY3RVJLUQX3DTPLKTDO.json","view_paper":"https://pith.science/paper/TWWCXSCQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.17544&json=true","fetch_graph":"https://pith.science/api/pith-number/TWWCXSCQY3RVJLUQX3DTPLKTDO/graph.json","fetch_events":"https://pith.science/api/pith-number/TWWCXSCQY3RVJLUQX3DTPLKTDO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TWWCXSCQY3RVJLUQX3DTPLKTDO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TWWCXSCQY3RVJLUQX3DTPLKTDO/action/storage_attestation","attest_author":"https://pith.science/pith/TWWCXSCQY3RVJLUQX3DTPLKTDO/action/author_attestation","sign_citation":"https://pith.science/pith/TWWCXSCQY3RVJLUQX3DTPLKTDO/action/citation_signature","submit_replication":"https://pith.science/pith/TWWCXSCQY3RVJLUQX3DTPLKTDO/action/replication_record"}},"created_at":"2026-07-05T07:39:38.428785+00:00","updated_at":"2026-07-05T07:39:38.428785+00:00"}