Uni-NLX: Unifying Textual Explanations for Vision and Vision-Language Tasks
This work addresses the inefficiency of task-specific models for generating textual explanations in vision and vision-language tasks, offering a more compact and unified solution.
The authors tackled the problem of training separate models for each natural language explanation (NLE) task by proposing Uni-NLX, a unified framework that consolidates seven NLE tasks into a single multi-task model, achieving comparable or better performance with 7x fewer parameters on 1M combined samples.
Natural Language Explanations (NLE) aim at supplementing the prediction of a model with human-friendly natural text. Existing NLE approaches involve training separate models for each downstream task. In this work, we propose Uni-NLX, a unified framework that consolidates all NLE tasks into a single and compact multi-task model using a unified training objective of text generation. Additionally, we introduce two new NLE datasets: 1) ImageNetX, a dataset of 144K samples for explaining ImageNet categories, and 2) VQA-ParaX, a dataset of 123K samples for explaining the task of Visual Question Answering (VQA). Both datasets are derived leveraging large language models (LLMs). By training on the 1M combined NLE samples, our single unified framework is capable of simultaneously performing seven NLE tasks including VQA, visual recognition and visual reasoning tasks with 7X fewer parameters, demonstrating comparable performance to the independent task-specific models in previous approaches, and in certain tasks even outperforming them. Code is at https://github.com/fawazsammani/uni-nlx