Search papers, labs, and topics across Lattice.
8
0
11
0
Current models falter on fine-grained facade elements, achieving only 33 IoU across geographic domains, highlighting a pressing need for better benchmarks.
UniGP reveals that joint training of controllable generation and dense prediction can significantly enhance performance without the need for complex designs, outperforming specialized models.
ClinHallu reveals that pinpointing the exact source of hallucinations in medical MLLM reasoning can drastically enhance model reliability.
State-of-the-art generative models struggle to maintain physical consistency and coherent interactions over time, revealing critical gaps in their world modeling capabilities.
K-Forcing accelerates token generation by 2.4-3.5x without abandoning the autoregressive backbone, making it a game-changer for high-load deployments.
Video diffusion models can achieve superior human motion control by leveraging 3D mesh tokenization, revealing a deeper understanding of 3D structures than previously thought.
Achieve high-fidelity video generation without compromising reasoning by progressively handing off generation from a lightweight, reasoning-aligned generator to a high-capacity pretrained generator in a shared latent space.
LLMs can be fine-tuned more efficiently by adapting experts in the frequency domain, leading to better performance with fewer parameters.
