ResearchResearch paperMultimodal Models · Large Language Models1 source · Oct 12, 2026

Teaching Vision-Language Models to Decline: Support Completion Against Mirage

Vision-language models answer questions about images they were never shown, a failure called mirage.

Such an answer can be correct without being justified by the evidence, and the difference matters wherever a model's output is checked by someone who did not see its input.
We locate the failure in the training objective: cross-entropy separates a model that reads the image from one that ignores it by only the conditional mutual information between answer and image, and no training example rewards declining.
The four-condition protocol on three base models, with seeds and component ablations, is in progress and not reported here.

[1]Teaching Vision-Language Models to Decline: Support Completion Against Mirage
arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 12, 09:21 AM
Vision-language models answer questions about images they were never shown, a failure called mirage.
Such an answer can be correct without being justified by the evidence, and the difference matters wherever a model's output is checked by someone who did not see its input.

Extractive summary: sentences quoted from the sources.

Before this