Jian Ge

DELOS is a novel contrastive-learning framework that efficiently and sensitively detects shallow, intermediate-to-long-period exoplanet transits in Kepler photometry, significantly outperforming traditional methods like BLS and TLS in low Signal-to-Noise Ratio regimes.

FLARE is a systematic conversion framework that enables a single checkpoint to support both autoregressive (AR) and diffusion-style parallel decoding for hybrid-attention large language models, achieving competitive performance and throughput gains.