Landscape of Thoughts: Visualizing the Reasoning Process of Large Language Models
Zhanke Zhou, Zhaocheng Zhu, Xuan Li, Mikhail Galkin, Xiao Feng, Sanmi Koyejo, Jian Tang, Bo Han
2025-04-02
Summary
This paper introduces a tool to help understand how AI models think step-by-step when solving problems.
What's the problem?
It's hard to know exactly how AI models arrive at their answers, making it difficult to improve them or ensure they're safe.
What's the solution?
The researchers created a visualization tool called 'landscape of thoughts' that shows the AI's reasoning process in a way that's easy to understand.
Why it matters?
This work matters because it can help researchers, developers, and users gain insights into AI decision-making, leading to better and more reliable AI systems.
Abstract
Numerous applications of large language models (LLMs) rely on their ability to perform step-by-step reasoning. However, the reasoning behavior of LLMs remains poorly understood, posing challenges to research, development, and safety. To address this gap, we introduce landscape of thoughts-the first visualization tool for users to inspect the reasoning paths of chain-of-thought and its derivatives on any multi-choice dataset. Specifically, we represent the states in a reasoning path as feature vectors that quantify their distances to all answer choices. These features are then visualized in two-dimensional plots using t-SNE. Qualitative and quantitative analysis with the landscape of thoughts effectively distinguishes between strong and weak models, correct and incorrect answers, as well as different reasoning tasks. It also uncovers undesirable reasoning patterns, such as low consistency and high uncertainty. Additionally, users can adapt our tool to a model that predicts the property they observe. We showcase this advantage by adapting our tool to a lightweight verifier that evaluates the correctness of reasoning paths. The code is publicly available at: https://github.com/tmlr-group/landscape-of-thoughts.