Visual CoT: Advancing Multi-Modal Language Models with a Comprehensive Dataset and Benchmark for Chain-of-Thought Reasoning
This addresses the problem of interpretability and handling high-resolution or small-region visual inputs in MLLMs for researchers and developers, though it is incremental as it builds on existing chain-of-thought methods.
The authors tackled the lack of interpretability and difficulty with complex visual inputs in Multi-Modal Large Language Models by introducing the Visual CoT dataset with 438k question-answer pairs and a multi-turn processing pipeline, resulting in improved performance as demonstrated through extensive experiments.
Multi-Modal Large Language Models (MLLMs) have demonstrated impressive performance in various VQA tasks. However, they often lack interpretability and struggle with complex visual inputs, especially when the resolution of the input image is high or when the interested region that could provide key information for answering the question is small. To address these challenges, we collect and introduce the large-scale Visual CoT dataset comprising 438k question-answer pairs, annotated with intermediate bounding boxes highlighting key regions essential for answering the questions. Additionally, about 98k pairs of them are annotated with detailed reasoning steps. Importantly, we propose a multi-turn processing pipeline that dynamically focuses on visual inputs and provides interpretable thoughts. We also introduce the related benchmark to evaluate the MLLMs in scenarios requiring specific local region identification. Extensive experiments demonstrate the effectiveness of our framework and shed light on better inference strategies. The Visual CoT dataset, benchmark, and pre-trained models are available on https://hao-shao.com/projects/viscot.html to support further research in this area.