Search papers, labs, and topics across Lattice.
Affiliation:
3
1
7
Robots can now learn complex manipulation tasks from a single demonstration video by recursively accessing fine-grained sub-event details, boosting success rates by up to 9%.
InfoMerge achieves an 85% reduction in visual tokens while retaining nearly all original performance, revolutionizing efficiency in Video-LLMs.
Compressing multi-dimensional human preferences into single binary labels severely degrades DPO training, but a semi-supervised approach can recover state-of-the-art performance without additional human annotation.
