Physics and EECS (AI track) double major at National Tsing Hua University, Taiwan, applying for PhD entry in 2027. At the HMI Lab I build generative decoders that reconstruct 3D spatial structure from fMRI. My other work is on predictive coding, associative memory and sampling-based inference. In 2026 I visited UCLA Samueli through the Taiwan Global Pathfinders program.
- 2nd of 609 teams, 2025 NSF HDR ML Challenge on anomaly detection, as team captain; 2nd in the gravitational-wave track
- 2nd worldwide, climate prediction track, 2026 NSF HDR ML Challenge on out-of-distribution generalization
- Presented our NSF HDR solution at an AAAI-25 workshop in Philadelphia
- Four submissions under review at ICLR 2027, one as first author
- iGEM 2023 Gold Medal, Grand Jamboree in Paris, dry lab lead
- 1st place, maker track, 2025 Mei-Chu Hackathon
I fix correctness defects in GPU, parallel computing, 3D and scientific Python libraries: wrong results, crashes and concurrency bugs. Each pull request carries a reproduction a maintainer can run and a test that fails before the fix and passes after it.
- X-Ray, chest X-ray pathology classification trained on one demographic subgroup and evaluated across all five, DenseNet121 against Xception
- NSF-HDR, team entry for the 2025 NSF HDR anomaly detection challenge, all three tracks
- quantum-allosteric-scanner, continuous time quantum walk prediction of allosteric sites from a single unbound protein structure
- MatrixQR, video generation that keeps a QR code scannable while it is stylised
- Arthur031221.github.io, bilingual research site, static HTML with no framework or runtime dependency

