{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:KJGDPNQRY3B7YE4WDUOXZED457","short_pith_number":"pith:KJGDPNQR","schema_version":"1.0","canonical_sha256":"524c37b611c6c3fc13961d1d7c907cefeae1bc2d7dcb129276446a677571329b","source":{"kind":"arxiv","id":"2305.06559","version":1},"attestation_state":"computed","paper":{"title":"Patch-wise Mixed-Precision Quantization of Vision Transformer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Junrui Xiao, Lianwei Yang, Qingyi Gu, Zhikai Li","submitted_at":"2023-05-11T04:34:10Z","abstract_excerpt":"As emerging hardware begins to support mixed bit-width arithmetic computation, mixed-precision quantization is widely used to reduce the complexity of neural networks. However, Vision Transformers (ViTs) require complex self-attention computation to guarantee the learning of powerful feature representations, which makes mixed-precision quantization of ViTs still challenging. In this paper, we propose a novel patch-wise mixed-precision quantization (PMQ) for efficient inference of ViTs. Specifically, we design a lightweight global metric, which is faster than existing methods, to measure the se"},"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":"2305.06559","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-05-11T04:34:10Z","cross_cats_sorted":[],"title_canon_sha256":"cd54783547a3ea1334edf3e5f3e0804472e0d465c47182f2c4b77a9df05dcea8","abstract_canon_sha256":"4069bf20eef9a2d54a2fadd57209a68979507685a3d032f8114dff22fc463fe3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:09:14.049600Z","signature_b64":"j8HCjpwCCYzfYy86ufs9ft/calcSF5ujPP/xPydRL1mRZDmVAQEkb9+yTOvMoijIeX7GLxLNxgUFZ0ssi9nQCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"524c37b611c6c3fc13961d1d7c907cefeae1bc2d7dcb129276446a677571329b","last_reissued_at":"2026-07-05T06:09:14.049200Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:09:14.049200Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Patch-wise Mixed-Precision Quantization of Vision Transformer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Junrui Xiao, Lianwei Yang, Qingyi Gu, Zhikai Li","submitted_at":"2023-05-11T04:34:10Z","abstract_excerpt":"As emerging hardware begins to support mixed bit-width arithmetic computation, mixed-precision quantization is widely used to reduce the complexity of neural networks. However, Vision Transformers (ViTs) require complex self-attention computation to guarantee the learning of powerful feature representations, which makes mixed-precision quantization of ViTs still challenging. In this paper, we propose a novel patch-wise mixed-precision quantization (PMQ) for efficient inference of ViTs. Specifically, we design a lightweight global metric, which is faster than existing methods, to measure the se"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.06559","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/2305.06559/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":"2305.06559","created_at":"2026-07-05T06:09:14.049257+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.06559v1","created_at":"2026-07-05T06:09:14.049257+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.06559","created_at":"2026-07-05T06:09:14.049257+00:00"},{"alias_kind":"pith_short_12","alias_value":"KJGDPNQRY3B7","created_at":"2026-07-05T06:09:14.049257+00:00"},{"alias_kind":"pith_short_16","alias_value":"KJGDPNQRY3B7YE4W","created_at":"2026-07-05T06:09:14.049257+00:00"},{"alias_kind":"pith_short_8","alias_value":"KJGDPNQR","created_at":"2026-07-05T06:09:14.049257+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/KJGDPNQRY3B7YE4WDUOXZED457","json":"https://pith.science/pith/KJGDPNQRY3B7YE4WDUOXZED457.json","graph_json":"https://pith.science/api/pith-number/KJGDPNQRY3B7YE4WDUOXZED457/graph.json","events_json":"https://pith.science/api/pith-number/KJGDPNQRY3B7YE4WDUOXZED457/events.json","paper":"https://pith.science/paper/KJGDPNQR"},"agent_actions":{"view_html":"https://pith.science/pith/KJGDPNQRY3B7YE4WDUOXZED457","download_json":"https://pith.science/pith/KJGDPNQRY3B7YE4WDUOXZED457.json","view_paper":"https://pith.science/paper/KJGDPNQR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.06559&json=true","fetch_graph":"https://pith.science/api/pith-number/KJGDPNQRY3B7YE4WDUOXZED457/graph.json","fetch_events":"https://pith.science/api/pith-number/KJGDPNQRY3B7YE4WDUOXZED457/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KJGDPNQRY3B7YE4WDUOXZED457/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KJGDPNQRY3B7YE4WDUOXZED457/action/storage_attestation","attest_author":"https://pith.science/pith/KJGDPNQRY3B7YE4WDUOXZED457/action/author_attestation","sign_citation":"https://pith.science/pith/KJGDPNQRY3B7YE4WDUOXZED457/action/citation_signature","submit_replication":"https://pith.science/pith/KJGDPNQRY3B7YE4WDUOXZED457/action/replication_record"}},"created_at":"2026-07-05T06:09:14.049257+00:00","updated_at":"2026-07-05T06:09:14.049257+00:00"}