Donglin Di

ArcAD reshapes cold-start anomaly detection by synthesizing pseudo-anomalies and clustering limited normal samples, leading to unprecedented performance gains.

Achieve nuanced control over portrait animations—even for subtle states like thinking or drowsiness—by using hierarchical agent planning to translate high-level labels into precise eye movements.

General-purpose vision-language models stumble when looking down from drones, but a new benchmark reveals that multi-task learning can substantially improve their aerial reasoning skills.

Ditch the optimization: MoRe achieves real-time 4D scene reconstruction from monocular video using a feedforward transformer that disentangles motion and structure.

By disentangling structure and motion in the latent space, CoWVLA achieves superior visuomotor learning compared to standard world-model and latent-action approaches.

VLMs can now excel at industrial anomaly detection by injecting domain-specific facts and aligning with expert preferences, achieving state-of-the-art zero-shot and one-shot performance.