Search papers, labs, and topics across Lattice.
4
0
9
SACHA achieves superior compression of 3D Gaussian head avatars by leveraging semantic awareness, resulting in both storage efficiency and high-quality rendering.
Removing gold answer strings from rewritten contexts can cause F1 scores to plummet by up to 64 points, underscoring their critical role in retrieval-augmented QA performance.
SpeechLLMs can be made significantly faster and more accurate at question answering by explicitly training their attention mechanisms to focus on relevant evidence.
