Search papers, labs, and topics across Lattice.
4
25
8
26
Ditch the static image: this method generates realistic talking avatars by learning from *videos* of the subject in completely different scenes.
Multi-level preference alignment in SignDPO significantly reduces semantic drift, outperforming traditional gloss-free models and challenging gloss-based benchmarks.
Forget dialogue summaries – FileGram builds user profiles directly from atomic file-system actions, unlocking a richer, more privacy-preserving approach to agent personalization.
Achieve surprisingly strong multimodal understanding and generation with a simple connector between off-the-shelf LLMs and diffusion models, using only a fraction of the parameters of larger models.
