Search papers, labs, and topics across Lattice.
4
0
7
0
Multimodal models can now handle audio natively with improved efficiency, achieving state-of-the-art results in complex tasks like document understanding and agentic computer use.
Noisy multimodal preference datasets are holding back reward model performance, but DT2IT-MRM offers a scalable curation strategy that achieves state-of-the-art results.
HeadRank achieves a remarkable 43-percentage-point improvement in selecting relevant documents from the middle-context zone, revolutionizing passage reranking efficiency.
Distributional divergence terms in Rate-Distortion-Perception theory, often used as a modeling principle, now have a theoretical justification rooted in synonymous sets.
