Search papers, labs, and topics across Lattice.
4
3
8
2
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.
Top LLMs can achieve medal-equivalent scores on elite science exams, but they falter on visual grounding and long-horizon consistency.
Kastor reduces forecasting error by nearly 43% while enhancing the physical fidelity of simulations, outperforming traditional methods in both accuracy and efficiency.
Achieving 1,500 tokens per second, DiffusionGemma redefines the speed-capability trade-off in language models, outpacing conventional autoregressive approaches.
