Search papers, labs, and topics across Lattice.
5
0
7
FusionVul outperforms traditional methods by integrating syntactic and structural insights, achieving unprecedented accuracy in detecting complex code vulnerabilities.
Gaze estimation models can now achieve comparable accuracy with 80-95% less labeled data, thanks to a semi-supervised approach that disentangles gaze components and learns robust representations via contrastive learning.
Achieve state-of-the-art sparse-view CT reconstruction with a diffusion model that generates visually consistent textures using fewer sampling steps, mitigating inherent randomness.
Ditch the RNNs and attention: PFGNet's frequency-guided gating achieves SOTA spatiotemporal prediction with a fully convolutional architecture, slashing parameters and FLOPs.
Diffusion models can learn surprisingly generalizable anatomical representations from unlabeled MRI data, enabling accurate multi-task diagnosis and segmentation across different joints and imaging conditions, even with limited labeled data.
