Search papers, labs, and topics across Lattice.
5
0
5
7
HyperSafe slashes harmful response rates in fine-tuned language models to below 1% without sacrificing task performance, revolutionizing safety alignment strategies.
Achieving encoding speeds up to 100,000 times faster than traditional methods, HyperGS redefines the efficiency of video representation while enhancing quality and generalization.
Over 1,100 submissions reveal groundbreaking advancements in sports video understanding, with new methods pushing the boundaries of action prediction and localization.
Attention sparsification via the $\alpha$-entmax transform can dramatically enhance CLIP's performance in open-vocabulary tasks by filtering out irrelevant context.
Hidden harmful supervision can infiltrate training data without detection, undermining the effectiveness of current safety measures.
