Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
6
0
7
23
Robots can achieve 70% success on complex bimanual tasks using only code for decision-making, bypassing runtime vision-language models entirely.
Most multimodal models fail to truly understand tool use in real-world egocentric videos, with even Gemini-3.1-Pro struggling to ground actions in visual evidence.
Current video generation models struggle with law-grounded reasoning, with the best achieving only 47% on the new Apple-PI benchmark.
Kinema4D unlocks zero-shot transfer in embodied AI by simulating physically plausible 4D robot-world interactions, moving beyond rigid 2D constraints.
Finally, a method exists to create 3D human-scene interaction models from casual captures that are stable enough for use in physics simulations and deployment on real-world robots.
Zero-shot synthesis of articulated human-object interactions is now possible by treating diffusion-generated videos as supervision for 4D scene reconstruction, unlocking physically grounded interactions beyond rigid manipulation.