Track synthesis, materials and catalysis as it is published.
Search papers, labs, and topics across Lattice.
6
0
9
5
A single adversarial token can bypass most safety filters.
Robotic recovery is now 74-93% faster by triggering corrections from action proposal disagreement and using efficient visual residuals, not slow VLM reasoning.
An 8B model can rival Gemini 3-Pro in generating detailed, temporally-aware scripts from long-form video, proving that targeted training trumps brute force scaling for narrative comprehension.
Generate diverse, physically plausible, and language-annotated whole-body motion data for humanoid robots at scale with this new interactive web-based pipeline.
Continuous diffusion can finally rival discrete methods in language modeling, thanks to LangFlow's novel architecture and training techniques.
A surprisingly simple VLA model, StarVLA-$\alpha$, beats more complex systems on real-world robotics tasks, suggesting that VLM backbones are more critical than intricate architectures.