Search papers, labs, and topics across Lattice.
3
0
5
AI-generated image detectors can be misled by spurious features, but DEAR prunes these distractions to enhance detection robustness against post-processing.
Skip the training and the hyperparameter tuning: Swift Sampling uses Taylor series to find the most informative frames in a video, beating existing methods with a 30x reduction in overhead.
Diffusion Transformers get a 3x speed boost without sacrificing image quality, thanks to a clever trick of dynamically adjusting patch sizes during denoising.
