1 link tagged with all of: multimodal + model-evaluation + automation + vlm + self-training
Links
This article outlines a method for training judges for Vision-Language Models (VLMs) without human annotations. The approach uses self-synthesized data in an iterative process to improve judgment accuracy, resulting in notable performance gains on various evaluation benchmarks.
vlm ✓
self-training ✓
model-evaluation ✓
multimodal ✓
automation ✓