Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
3
4
6
12
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.
FATE not only retains temporal information but also encodes synchronization in a reusable embedding space, outperforming existing models in both semantic and temporal tasks.
DSpark shifts the performance landscape of LLM inference, achieving up to 85% faster generation speeds while maintaining high throughput.