Concept文本捷径问题
Text Shortcut
A phenomenon in multimodal evaluation in which a model can answer correctly using only cues in the question text without genuinely understanding the image, leading to overestimation of multimodal capability.
Timeline (last 90 days)
Oct 23
Many multimodal benchmarks evaluate models by looking only at final answer accuracy, ignoring whether the model actually relies on images; a common problem is that textual shortcuts lead to overestimation of multimodal capabilities.
Unverified50%