Search papers, labs, and topics across Lattice.
Affiliation:
3
0
8
0
Embodied agents improve on complex tasks by recombining learned skills, but their ability to extract value from additional experience saturates, failing to scale with learning samples.
Natural-language instructions can now handle both zero-shot speech synthesis and surgical acoustic editing within a single unified model, operating at 4-step distilled inference speeds without classifier-free guidance.
Directly optimizing internal reasoning from outcome feedback, GradCuit achieves a 6.6% accuracy boost over chain-of-thought prompting while enhancing robustness and interpretability.
