Search papers, labs, and topics across Lattice.
4
901
6
9
Atomic visual perception in MLLMs is largely unsolved, with no model surpassing 60% accuracy on a new benchmark designed to isolate perceptual capabilities.
Kimi K3's innovative architecture achieves a 2.5x scaling efficiency improvement, enabling robust performance across diverse long-horizon tasks.
Forget fixed residual connections: Attention Residuals let each layer selectively attend to previous layers, boosting performance and gradient flow in deep LLMs.
Open-source multimodal models just leveled up: InternVL3 rivals closed-source titans like GPT-4o by pre-training vision and language together from the start.
