ShapeWorld - A new test methodology for multimodal language understanding
This provides a controlled test methodology for researchers in multimodal AI, though it is incremental as it builds on existing evaluation approaches.
The authors tackled the problem of evaluating multimodal deep learning models by introducing ShapeWorld, a framework that generates artificial data to test language understanding and generalization, showing detailed insights into model capabilities and limitations across four visual question answering tasks.
We introduce a novel framework for evaluating multimodal deep learning models with respect to their language understanding and generalization abilities. In this approach, artificial data is automatically generated according to the experimenter's specifications. The content of the data, both during training and evaluation, can be controlled in detail, which enables tasks to be created that require true generalization abilities, in particular the combination of previously introduced concepts in novel ways. We demonstrate the potential of our methodology by evaluating various visual question answering models on four different tasks, and show how our framework gives us detailed insights into their capabilities and limitations. By open-sourcing our framework, we hope to stimulate progress in the field of multimodal language understanding.