VisDoT : Enhancing Visual Reasoning through Human-Like Interpretation Grounding and Decomposition of Thought

TL;DR

VisDoT enhances visual reasoning in vision-language models by grounding perception in human-like interpretation and decomposing questions into perception and logic, leading to significant performance improvements on chart-based and open-domain VQA benchmarks.

Contribution

The paper introduces VisDoT, a novel framework that formalizes perceptual tasks and employs Decomposition-of-Thought prompting to improve visual reasoning in LVLMs.

Findings

+11.2% improvement on ChartQA with fine-tuning

Surpasses GPT-4o on ChartQAPro benchmark

+33.2% improvement on VisDoTQA benchmark

Abstract

Large vision-language models (LVLMs) struggle to reliably detect visual primitives in charts and align them with semantic representations, which severely limits their performance on complex visual reasoning. This lack of perceptual grounding constitutes a major bottleneck for chart-based reasoning. We propose VisDoT, a framework that enhances visual reasoning through human-like interpretation grounding. We formalize four perceptual tasks based on the theory of graphical perception, including position and length. Building on this foundation, we introduce Decomposition-of-Thought (DoT) prompting, which sequentially separates questions into visual perception sub-questions and logic sub-questions. Fine-tuning InternVL with VisDoT achieves a +11.2% improvement on ChartQA and surpasses GPT-4o on the more challenging ChartQAPro benchmark. On the newly introduced VisDoTQA benchmark, the model improves by +33.2%. Furthermore, consistent zero-shot gains on diverse open-domain VQA benchmarks confirm the generalizability of the perception-logic separation strategy for visual question answering. VisDoT leverages human-like perception to enhance visual grounding, achieving state-of-the-art chart understanding and interpretable visual reasoning.