Search papers, labs, and topics across Lattice.
University of Hong Kong, ByteDance Seed
2
0
6
Ditch the VAE bottleneck: Representation Forcing lets you train unified multimodal models to generate high-quality images directly from pixels, rivaling VAE-based approaches without the architectural constraint.
LLMs can help novice programmers build social robots, but only if the AI scaffolds the design process instead of taking over.
