3 papers indexed here
We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.
Not the right person? Other researchers publish under this name.
Across three vision-language model architectures (LLaVA-1.5-7B, Qwen2.5-VL-7B, InternVL3-8B), we report a universal negative finding for mid-layer interpretability. On POPE -- the benchmark common to all three -- the mid layers encode the ground-truth answer in 68-91% of errors, yet this signal is not causally active f...
No configuration the authors test improves both the compositional corner and aggregate accuracy, and the rare-input long tail thus points to architectural change rather than added capacity.
VSI is not a universal best abstention signal; it is a sample-intrinsic indicator of vision-ignoring failure, best used as a conditional ensemble component.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
