We present ASTERIS, an astronomical self-supervised transformer-based denoising algorithm that integrates spatiotemporal information across multiple exposures. ASTERIS improves detection limits by 1.0 magnitude at 90% completeness and purity while preserving the point spread function and photometric accuracy. Applied to deep JWST images, it identifies three times more redshift ≳9 galaxy candidates, with rest-frame UV luminosity 1.0 magnitude fainter than previous methods.

We develop a light-field-based wide-field wavefront sensor (WWS) achieving direct observation of atmospheric turbulence over 1,100 arcsec at 30 Hz, with turbulence dynamics predicted 33 ms ahead. Described by MIT Prof. Dirk Englund as the state-of-the-art real-time wavefront sensing chip.

We propose a meta-imaging sensor achieving gigapixel photography with a single spherical lens, enabling multisite aberration correction across 1,000 arcseconds on an 80-cm ground-based telescope. Recognized as one of China's Top 10 Optical Breakthroughs.