Search papers, labs, and topics across Lattice.
1
0
2
MLLMs generate critiques that are more verbose and less selective than human reviewers, raising questions about the validity of traditional evaluation metrics.
