Towards Reliable VLM Judges: State-Conditional Invariance and Presentation-Aware Diagnostics
Vision-language models (VLMs) are increasingly used as automated judges for multimodal and embodied tasks, yet aggregate accuracy or consistency alone does not reveal whether a judge changes its verdict for the right reason. We propose State-Conditional Invariance (SCI), a reliability principle requiring a judge to remain stable under semantics-preserving changes in presentation while responding to causal interventions that change the task outcome. We instantiate SCI in GridWM-Judge, a simulator-grounded diagnostic benchmark built from deterministic MiniGrid trajectories with aligned Full, evidence-removal, and counterfactual variants, plus presentation and representation probes. Across five diagnostic tasks, GridWM-Judge tests local perception, action-conditioned transitions, presentation stability, representation exchangeability, and Full-CF causal discrimination. Our diagnostic evaluations reveal heterogeneous failure modes, including sensitivity to evidence presentation and encoding, low activation on successful trajectories, and weak paired counterfactual discrimination. Perception-controlled analyses further indicate that visual grounding and objective recovery are…
Reliability framework for VLM judges with state-conditional invariance diagnostics.
- arXiv
- 2611.13596
- Authors
- Qinan Zhang, Xinyu Wang, Qihang Jin, Tianyuan Wang, Hongbing Luo and 4 more
- Categories
- cs.AI · cs.CL · cs.CV · cs.LG