Diffuse to Detect: Generative Diffusion Models for Unsupervised IC Anomaly Detection
Yuxuan Yin 2 , Chen He 3 , Todd Jacobs 3 , Jialei He 3 , Boxun Xu 2 , Robert Jin 3 , Peng Li 2 · 2026
Read Diffuse to Detect: Generative Diffusion Models for Unsupervised IC Anomaly Detection by Yuxuan Yin 2 , Chen He 3 , Todd Jacobs 3 , Jialei He 3 , Boxun Xu 2 , Robert Jin 3 , Peng Li 2 (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-26468.txt.
Description
Yuxuan Yin and six colleagues screen latent chip defects without labeled failures using a diffusion transformer. Raw test readings first pass through an autoencoder, then reshape into tokens enriched with per-device wafer-position data and sinusoidal cues. Scores come from noise-prediction error over mid-range diffusion steps, enabling fast wafer-scale screening with no manual feature work. On industrial 16-nanometer chip test data under extreme class imbalance, the method reaches state-of-the-art performance. Latent-space reconstruction residuals also point to where failures sit.
Catalog record
| CREATOR | Yuxuan Yin 2 , Chen He 3 , Todd Jacobs 3 , Jialei He 3 , Boxun Xu 2 , Robert Jin 3 , Peng Li 2 |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43632 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.26468