Corruption-Risk Reversal in Adaptive Visual Evidence Routing for Medical Vision–Language Models

Adaptive visual computation is usually motivated as an efficiency problem: additional visual tokens are spent only when a global view is insufficient. For an autoregressive vision–language model (VLM), however, an acquired crop is not simply more information. It changes both the visible evidence and the generation context, and can therefore overturn a correct global answer. This reliability problem is formulated as cost-sensitive evidence routing among answering, acquiring a view, and abstaining. The formulation separates the availability of a better action from the router's ability to select it, and decomposes acquisition value into helpful corrections and harmful corruptions. Empirically, every policy is reconstructed from immutable paired counterfactual outputs with a shared global response, exact-image grouping, revision-pinned models, and policies frozen before transfer. On historically exposed M3CoTBench development, signed-utility routing raises MedGemma-4B accuracy from 24.55% to 30.91% ($+6.36\text{pp}$; 95% image-group bootstrap CI $[+2.59,+11.21]$). Under locked PathVQA shift, the same rule lowers accuracy from 70.31% to 63.67% ($-6.64\text{pp}$; CI $[-10.55,-2.73]$). A separately frozen confidence-ranked plurality policy likewise lowers Qwen accuracy on 2,996 prospective 3MDBench cases from 40.92% to 39.05%, correcting 16 answers and corrupting 72. On DDI, a conservative unanimity gate avoids added errors only by making no replacements, while global accuracy conceals 0.58% malignant sensitivity. Across these evaluations, the observed pattern is termed corruption-risk reversal: uncertainty continues to identify fragile cases, but ceases to distinguish acquisitions that repair an answer from those that damage it. These results establish neither a universal failure of cropping nor clinical safety. They instead show that efficiency claims alone are insufficient without action headroom, transferable harm estimation, safe aggregation, and nontrivial coverage.