Phong Nguyen

Recency Forcing is proposed, which applies a non-positive, timestep-dependent bias, termed Temporal Response Bias (TRB), on pre-softmax attention logits derived directly from $R, closing the train-inference gap without modifying context length or training objectives.

Compact diffusion models can now leverage the power of high-capacity Teachers without architectural changes, achieving remarkable performance gains.

Optimizing threading in RTAC cuts inference latency and boosts control responsiveness, making robotic manipulation viable for low-cost agricultural applications.

Weak-to-strong reward models can ace the test but still fail in the real world, revealing a hidden brittleness in current preference learning approaches.

Forget waiting a minute for garment generation: SwiftTailor slashes inference times while boosting accuracy by representing 3D garments as geometry images.