Do Multimodal Language Models Really Understand Direction? A Benchmark for Compass Direction Reasoning
This work addresses the underexplored problem of compass direction reasoning for intelligent systems, though it is incremental as it builds on existing spatial reasoning benchmarks.
The authors tackled the problem of evaluating compass direction reasoning in multimodal language models by proposing the Compass Direction Reasoning (CDR) benchmark, revealing that most models perform at random guessing levels, but mixdata and CoT fine-tuning methods significantly enhance performance.
Direction reasoning is essential for intelligent systems to understand the real world. While existing work focuses primarily on spatial reasoning, compass direction reasoning remains underexplored. To address this, we propose the Compass Direction Reasoning (CDR) benchmark, designed to evaluate the direction reasoning capabilities of multimodal language models (MLMs). CDR includes three types images to test spatial (up, down, left, right) and compass (north, south, east, west) directions. Our evaluation reveals that most MLMs struggle with direction reasoning, often performing at random guessing levels. Experiments show that training directly with CDR data yields limited improvements, as it requires an understanding of real-world physical rules. We explore the impact of mixdata and CoT fine-tuning methods, which significantly enhance MLM performance in compass direction reasoning by incorporating diverse data and step-by-step reasoning, improving the model's ability to understand direction relationships.