State Beyond Appearance: Diagnosing and Improving State Consistency in Dial-Based Measurement Reading

TL;DR

This paper identifies the limitations of current multimodal large language models in dial-based measurement reading, diagnosing their reliance on superficial cues and proposing a new framework, TriSCA, to improve state consistency.

Contribution

The paper introduces TriSCA, a tri-level alignment framework that enhances state consistency in dial-based measurement reading for multimodal models.

Findings

Current MLLMs perform poorly on dial-based tasks under viewpoint and illumination changes.

Probing reveals MLLMs do not preserve local structure of dial states.

TriSCA significantly improves accuracy and robustness on benchmarks.

Abstract

Multimodal large language models (MLLMs) have achieved impressive progress on general multimodal tasks, yet they remain brittle on dial-based measurement reading. In this paper, we study this problem through controlled benchmarks and feature-space probing, and show that current MLLMs not only achieve unsatisfactory accuracy on dial-based readout, but also suffer sharp performance drops under viewpoint and illumination changes even when the underlying dial state remains fixed. Our probing analysis further reveals that same-state samples under appearance variation are not consistently clustered, while neighboring states fail to preserve the local structure implied by continuous dial values. These findings suggest that existing MLLMs largely ignore the intrinsic state geometry of dial measurement tasks and instead rely on superficial appearance cues. Motivated by this diagnosis, we propose TriSCA, a tri-level state-consistent alignment framework for dial-based measurement reading. Specifically, TriSCA consists of state-distance-aware representation alignment, metadata-grounded observation-to-state supervision, and state-aware objective alignment. Extensive ablation studies and evaluation experiments on controlled clock and gauge benchmarks, together with evaluation on an external real-world benchmark, demonstrate the effectiveness of our method.