Cognitive Mismatch in Multimodal Large Language Models for Discrete Symbol Understanding
This work addresses a critical gap in AI capabilities for scientific and abstract domains, exposing a cognitive mismatch that hinders true understanding of symbolic languages.
The paper tackled the problem of Multimodal Large Language Models (MLLMs) struggling to process discrete symbols like mathematical formulas and chemical structures, revealing that models often fail at basic symbol recognition but succeed in complex reasoning tasks, indicating reliance on linguistic probability rather than true visual perception.
While Multimodal Large Language Models (MLLMs) have achieved remarkable success in interpreting natural scenes, their ability to process discrete symbols -- the fundamental building blocks of human cognition -- remains a critical open question. Unlike continuous visual data, symbols such as mathematical formulas, chemical structures, and linguistic characters require precise, deeper interpretation. This paper introduces a comprehensive benchmark to evaluate how top-tier MLLMs navigate these "discrete semantic spaces" across five domains: language, culture, mathematics, physics, and chemistry. Our investigation uncovers a counterintuitive phenomenon: models often fail at basic symbol recognition yet succeed in complex reasoning tasks, suggesting they rely on linguistic probability rather than true visual perception. By exposing this "cognitive mismatch", we highlight a significant gap in current AI capabilities: the struggle to truly perceive and understand the symbolic languages that underpin scientific discovery and abstract thought. This work offers a roadmap for developing more rigorous, human-aligned intelligent systems.