Search papers, labs, and topics across Lattice.
3
0
6
5
Encoder-decoders are out: a decoder-only architecture for large view synthesis not only slashes parameters but also beats the state-of-the-art, even outperforming per-scene optimized 3DGS in some cases.
A simple resampling strategy closes the "Thinking-Acting Gap" in agentic VLMs, enabling smaller models to outperform larger ones on multimodal reasoning tasks.
Time is a learnable visual concept: models can now reason about and manipulate the flow of time in videos, opening doors to temporally controllable video generation and temporal forensics.
