Search papers, labs, and topics across Lattice.
1
0
3
Visual reasoning in multimodal models is highly variable, with accuracy plummeting over 10 points when feedback is corrupted, revealing hidden dependencies on visual states.
