Search papers, labs, and topics across Lattice.
Affiliation:
4
0
8
6
Free-form language reasoning can dramatically enhance robotic manipulation, outperforming traditional instruction-based methods in complex tasks.
Different views of the same problem reveal hidden reasoning paths, enabling VLMs to achieve unprecedented accuracy in multimodal reasoning tasks.
A 4B model can rival the mathematical reasoning of models 30x its size, proving that clever training trumps brute force scaling.
Forget simple scaling laws: the compute-optimal number of parallel rollouts in LLM RL plateaus, revealing distinct mechanisms for easy vs. hard problems.
