Analyzing ancient manuscripts has traditionally been a labor-intensive and time-consuming task for philologists. While recent advancements in LMMs have demonstrated their potential across diverse domains, their effectiveness in manuscript study remains underexplored. In this paper, we introduce MS-Bench, the first comprehensive benchmark co-developed with archaeologists, comprising 5,076 high-resolution images from 4th to 14th century and 9,982 expert-curated questions across nine sub-tasks aligned with archaeological workflows. Through four prompting strategies, we systematically evaluate 32 LMMs on their effectiveness, robustness, and cultural contextualization. Our analysis reveals scale-driven performance and reliability improvements, prompting strategies' impact on performance (CoT has two-sides effect, while visual retrieval-augmented prompts provide consistent boost), and task-specific preferences depending on LMM’s visual capabilities. Although current LMMs are not yet capable of replacing domain expertise, they demonstrate promising potential to accelerate manuscript research through future human–AI collaboration.

Scratch testing is an effective technique for evaluating surface mechanical properties of materials. However, for motor-driven scratch testers, the micron-scale axial transmission clearance of ball screws and the millisecond response time of the motor greatly affect the compensation of the normal force, leading to a large fluctuation of the normal force. Accordingly, this study proposed a compliant mechanism adaptive suppression method to adapt to the change in geometric characteristics and mechanical properties of a sample surface, thereby reducing the fluctuation of the normal force. The structural characteristics of the compliant mechanism were discussed, followed by theoretical and simulation analysis to demonstrate the effectiveness of the method. Comparative experiments were performed on materials with heterogeneous mechanical properties and rough surfaces by scratch testers with and without the compliant mechanism. The results confirmed that the proposed method could well suppress the fluctuation of the normal force in motor-driven scratch testers under various experimental conditions.