Search papers, labs, and topics across Lattice.
5
0
8
One-step diffusion models can achieve state-of-the-art image quality with only 10% of the parameters by focusing on weight direction changes during distillation.
BabelRS disentangles modality alignment from downstream task learning in multi-modal remote sensing, leading to more stable training and improved detection accuracy.
Ditch the deep thought: this new agentic search framework slashes reasoning steps by 70% while boosting accuracy by prioritizing parallel evidence gathering.
Knowledge Graph Completion gets a boost: KGT's dedicated entity tokens and decoupled prediction heads let LLMs reason about KGs without being constrained by token fragmentation.
Forget retraining: OmniVTON++ achieves state-of-the-art virtual try-on across diverse datasets and garment types using a completely training-free approach.
