Thinking with Images for Multimodal Reasoning: Foundations, Methods, and Future Frontiers

TL;DR

This survey explores the evolution of multimodal reasoning in AI from static image understanding to dynamic, thinking-with-images paradigms, highlighting foundational principles, methods, benchmarks, and future challenges.

Contribution

It establishes a three-stage framework for think-with-image AI, reviews core methods, analyzes benchmarks, and outlines future research directions.

Findings

Introduction of a three-stage evolution framework

Comprehensive review of core methods at each stage

Analysis of benchmarks and future challenges

Abstract

Recent progress in multimodal reasoning has been significantly advanced by textual Chain-of-Thought (CoT), a paradigm where models conduct reasoning within language. This text-centric approach, however, treats vision as a static, initial context, creating a fundamental "semantic gap" between rich perceptual data and discrete symbolic thought. Human cognition often transcends language, utilizing vision as a dynamic mental sketchpad. A similar evolution is now unfolding in AI, marking a fundamental paradigm shift from models that merely think about images to those that can truly think with images. This emerging paradigm is characterized by models leveraging visual information as intermediate steps in their thought process, transforming vision from a passive input into a dynamic, manipulable cognitive workspace. In this survey, we chart this evolution of intelligence along a trajectory of increasing cognitive autonomy, which unfolds across three key stages: from external tool exploration, through programmatic manipulation, to intrinsic imagination. To structure this rapidly evolving field, our survey makes four key contributions. (1) We establish the foundational principles of the think with image paradigm and its three-stage framework. (2) We provide a comprehensive review of the core methods that characterize each stage of this roadmap. (3) We analyze the critical landscape of evaluation benchmarks and transformative applications. (4) We identify significant challenges and outline promising future directions. By providing this structured overview, we aim to offer a clear roadmap for future research towards more powerful and human-aligned multimodal AI.