Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
4
3
9
4
The DeepSeek-V4.1-Flash model, a multimodal Mixture-of-Experts model with 552B backbone parameters and support for contexts of up to one million tokens, is introduced, substantially improving cost efficiency for agentic workloads and pushing the limits of KV cache compression.
TWLA achieves unprecedented 1.58-bit weight and 4-bit activation quantization while preserving accuracy, revolutionizing the deployment of large language models.
Pathologists using AI assistance can boost diagnostic accuracy by 8% and slash diagnostic time by 20% in lung pathology, according to a prospective, multi-center RCT.
MoBiE slashes the cost of running Mixture-of-Experts models, achieving 2x speedups and massive accuracy gains by intelligently binarizing weights *without* adding storage overhead.