Search papers, labs, and topics across Lattice.
Affiliation:
4
0
8
2
Treating robot skill calls as runtime-verified execution proposals rather than blind action predictions enables self-correcting VLA autonomy that reaches 97.4% success on LIBERO.
PaperGym achieves a remarkable 73.48 on ResearchQA, outperforming larger models and redefining how AI can generate and evaluate research plans.
Even the best multimodal LLMs are surprisingly bad at understanding and remembering the "self" in egocentric video, lagging human performance by 40-50% on personalized question answering.
LLM agents can internalize skills via in-context RL, achieving zero-shot autonomous behavior without the token overhead and retrieval noise of traditional methods.
