{"paper":{"title":"UniPCB: A Generation-Assisted Detection Framework for PCB Defect Inspection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A joint generation-detection framework for PCBs uses multi-modal synthesis to augment scarce defect data and reach 98.0 percent mAP@0.5.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Huanqi Wu, Huan Zhang, Jiangzhong Cao, Lianghong Tan, Linwei Zhu, Xu Zhang, Yichu Xu, Zishan Su","submitted_at":"2026-05-06T08:30:27Z","abstract_excerpt":"In the Industrial Internet of Things (IIoT), enabling intelligent, real-time Printed Circuit Board (PCB) defect inspection is critical for ensuring product reliability. However, existing IIoT-based visual inspection systems face two compounding challenges: scarce and imbalanced defect samples that limit model training, and insufficient feature representation under complex circuit backgrounds. Existing generation methods rely on single-modality conditions with coarse structural control, while detection methods improve architectures without addressing the data bottleneck. To resolve both challen"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"UniPCB achieves mAP@0.5 of 98.0% and mAP@0.5:0.95 of 61.8% on defect detection, surpassing all compared methods, while the generation branch attains an FID of 129.61 and SSIM of 0.619, outperforming existing conditional generation approaches.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The synthesized defect samples are realistic and distributionally aligned enough with real IIoT data that adding them improves detection performance without introducing artifacts or domain shift that harms real-world accuracy.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"UniPCB reaches 98.0% mAP@0.5 on PCB defect detection by synthesizing realistic defects via multi-modal diffusion and feeding them into an attention-based detector.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A joint generation-detection framework for PCBs uses multi-modal synthesis to augment scarce defect data and reach 98.0 percent mAP@0.5.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"fc88ad2ea8d830c5d22d459d838a46e8030251af4be8f66dbe568209046e8c27"},"source":{"id":"2605.04635","kind":"arxiv","version":3},"verdict":{"id":"00257ef7-893d-436b-9d58-31a055f543b7","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-12T00:46:45.340438Z","strongest_claim":"UniPCB achieves mAP@0.5 of 98.0% and mAP@0.5:0.95 of 61.8% on defect detection, surpassing all compared methods, while the generation branch attains an FID of 129.61 and SSIM of 0.619, outperforming existing conditional generation approaches.","one_line_summary":"UniPCB reaches 98.0% mAP@0.5 on PCB defect detection by synthesizing realistic defects via multi-modal diffusion and feeding them into an attention-based detector.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The synthesized defect samples are realistic and distributionally aligned enough with real IIoT data that adding them improves detection performance without introducing artifacts or domain shift that harms real-world accuracy.","pith_extraction_headline":"A joint generation-detection framework for PCBs uses multi-modal synthesis to augment scarce defect data and reach 98.0 percent mAP@0.5."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.04635/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-20T11:36:29.937952Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T22:31:19.938568Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T14:17:33.642779Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"4da824751a78e09298c87e1e5adb36d3691093a34f995a95a02f87eb5516b026"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"8faa1a2e374e1f782e6e5c5490b3262889673dcfa59374db019c60185c8747d7"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}