Sina Farsiu

Stop training your image restoration models to mimic flawed ground truth; instead, explicitly optimize for perceptual quality using a plug-and-play module guided by No-Reference Image Quality Assessment.

Instead of discarding noisy pseudo-labels in image restoration, QualiTeacher leverages them by teaching the model to understand and even surpass the quality levels they represent.

Forget massive datasets: PRIMA leverages risk-integrated image-metadata alignment to achieve state-of-the-art medical diagnosis with LLMs, proving that smart pre-training can trump brute force.