Align Anything: Training All-Modality Models to Follow Instructions with Language Feedback

Open in new window