Search papers, labs, and topics across Lattice.
8
0
11
Transforming video generation from a pixel sampling problem to a structured orchestration of the physical world, WNM enables unprecedented control and efficiency in content creation.
Achieving a 90% boost in prefill throughput for MoE models could redefine the efficiency of large-scale language model serving.
TEASR achieves seamless any-step sampling and outperforms traditional methods, all while eliminating the need for bulky auxiliary models.
Large-scale generative models struggle with low-level vision tasks, revealing critical performance gaps that conventional metrics fail to capture.
Current video generation benchmarks overlook crucial aspects of physical plausibility and temporal coherence, highlighting the need for holistic evaluation metrics like PhyScore.
Software vulnerability detection gets a serious upgrade: aligning code with developer comments boosts F1 scores by up to 27% compared to traditional code-only methods.
Achieve state-of-the-art dynamic 3D reconstruction from sparse views by intelligently fusing explicit 3D geometry with generative priors, avoiding common pitfalls like structural drift and temporal inconsistency.
Finally, a blind face restoration method that doesn't just hallucinate details, but lets you precisely control facial attributes via text prompts while maintaining high fidelity.
