A large-scale randomized study of large language model feedback in peer review - Nature
In a randomized controlled study at ICLR 2025, Thakkar et al. demonstrate that large language model-generated feedback can make reviews more informative while enhancing reviewer–author engagement.
Comments