Search papers, labs, and topics across Lattice.
Department of Computer Science, University of Wisconsin-Madison
4
0
10
2
OAT achieves up to 5000 times faster failure attribution for LLM agents without the need for costly step-level supervision.
Progress advantage reveals a powerful, annotation-free scoring mechanism that outperforms traditional reward models in LLM agentic settings.
Discovering when a robot's about to fail just got easier: Hide-and-Seek pinpoints failure signals in VLA trajectories using only coarse, trajectory-level labels, ditching the need for expensive step-by-step annotations.
Forget tweaking prompts – understanding how retrieved context warps an LLM's hidden states is the key to unlocking better RAG performance.
