Foundation of Intelligence: Review of Math Word Problems from Human Cognition Perspective

2025-10-24

AI Summary

Prior mathematical word problem (MWP) research lacks a systematic taxonomy and integration of cognitive perspectives. Method: Grounded in cognitive science, this work identifies five core human competencies—problem understanding, logical organization, associative memory, critical thinking, and knowledge acquisition—and constructs a unified classification framework and evaluation protocol. It conducts a decade-long survey of methodological evolution, performing cross-paradigm comparative analysis across rule-based systems, neural solvers, and large language models, with standardized reimplementation and evaluation on five mainstream benchmarks. Contribution/Results: The paper releases an open-source repository—including code, evaluation results, and analytical insights—significantly enhancing reproducibility, comparability, and cognitive interpretability in MWP research. This establishes a novel paradigm for modeling and evaluating AI’s reasoning capabilities.

Technology Category

• Knowledge Representation and Reasoning: Computational Complexity of Reasoning
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning
• Planning, Routing, and Scheduling: Model-Based Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Math word problem (MWP) serves as a fundamental research topic in artificial intelligence (AI) dating back to 1960s. This research aims to advance the reasoning abilities of AI by mirroring the human-like cognitive intelligence. The mainstream technological paradigm has evolved from the early rule-based methods, to deep learning models, and is rapidly advancing towards large language models. However, the field still lacks a systematic taxonomy for the MWP survey along with a discussion of current development trends. Therefore, in this paper, we aim to comprehensively review related research in MWP solving through the lens of human cognition, to demonstrate how recent AI models are advancing in simulating human cognitive abilities. Specifically, we summarize 5 crucial cognitive abilities for MWP solving, including Problem Understanding, Logical Organization, Associative Memory, Critical Thinking, and Knowledge Learning. Focused on these abilities, we review two mainstream MWP models in recent 10 years: neural network solvers, and LLM based solvers, and discuss the core human-like abilities they demonstrated in their intricate problem-solving process. Moreover, we rerun all the representative MWP solvers and supplement their performance on 5 mainstream benchmarks for a unified comparison. To the best of our knowledge, this survey first comprehensively analyzes the influential MWP research of the past decade from the perspective of human reasoning cognition and provides an integrative overall comparison across existing approaches. We hope it can inspire further research in AI reasoning. Our repository is released on https://github.com/Ljyustc/FoI-MWP.

Problem

• Lacks systematic taxonomy for math word problem surveys
• Needs comprehensive review of MWP research through human cognition lens
• Requires unified performance comparison of existing MWP solving approaches

Innovation

• Reviewing MWP solvers via human cognition abilities
• Comparing neural network and LLM based solvers
• Providing unified performance comparison on benchmarks