UNK-VQA: A Dataset and a Probe into the Abstention Ability of Multi-modal Large Models
This addresses the need for trustworthy AI systems by providing a benchmark for improving abstention capabilities in VQA models, though it is incremental as it builds on existing VQA research with a new dataset and evaluation.
The paper tackles the problem of teaching Visual Question Answering (VQA) models to abstain from answering unanswerable questions by introducing the UNK-VQA dataset, which uses deliberate perturbations to create challenging unanswerable cases, and reveals significant limitations in zero- and few-shot performance of multi-modal large models on this dataset.
Teaching Visual Question Answering (VQA) models to refrain from answering unanswerable questions is necessary for building a trustworthy AI system. Existing studies, though have explored various aspects of VQA but somewhat ignored this particular attribute. This paper aims to bridge the research gap by contributing a comprehensive dataset, called UNK-VQA. The dataset is specifically designed to address the challenge of questions that models do not know. To this end, we first augment the existing data via deliberate perturbations on either the image or question. In specific, we carefully ensure that the question-image semantics remain close to the original unperturbed distribution. By this means, the identification of unanswerable questions becomes challenging, setting our dataset apart from others that involve mere image replacement. We then extensively evaluate the zero- and few-shot performance of several emerging multi-modal large models and discover their significant limitations when applied to our dataset. Additionally, we also propose a straightforward method to tackle these unanswerable questions. This dataset, we believe, will serve as a valuable benchmark for enhancing the abstention capability of VQA models, thereby leading to increased trustworthiness of AI systems. We have made the dataset (https://github.com/guoyang9/UNK-VQA) available to facilitate further exploration in this area.