Fast, Slow, and Tool-augmented Thinking for LLMs: A Review

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Jia, Xinda, Li, Jinpeng, Wang, Zezhong, Li, Jingjing, Zeng, Xingshan, Wang, Yasheng, Zhang, Weinan, Yu, Yong, Liu, Weiwen
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915448689262592
author Jia, Xinda
Li, Jinpeng
Wang, Zezhong
Li, Jingjing
Zeng, Xingshan
Wang, Yasheng
Zhang, Weinan
Yu, Yong
Liu, Weiwen
author_facet Jia, Xinda
Li, Jinpeng
Wang, Zezhong
Li, Jingjing
Zeng, Xingshan
Wang, Yasheng
Zhang, Weinan
Yu, Yong
Liu, Weiwen
contents Large Language Models (LLMs) have demonstrated remarkable progress in reasoning across diverse domains. However, effective reasoning in real-world tasks requires adapting the reasoning strategy to the demands of the problem, ranging from fast, intuitive responses to deliberate, step-by-step reasoning and tool-augmented thinking. Drawing inspiration from cognitive psychology, we propose a novel taxonomy of LLM reasoning strategies along two knowledge boundaries: a fast/slow boundary separating intuitive from deliberative processes, and an internal/external boundary distinguishing reasoning grounded in the model's parameters from reasoning augmented by external tools. We systematically survey recent work on adaptive reasoning in LLMs and categorize methods based on key decision factors. We conclude by highlighting open challenges and future directions toward more adaptive, efficient, and reliable LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2508_12265
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Fast, Slow, and Tool-augmented Thinking for LLMs: A Review
Jia, Xinda
Li, Jinpeng
Wang, Zezhong
Li, Jingjing
Zeng, Xingshan
Wang, Yasheng
Zhang, Weinan
Yu, Yong
Liu, Weiwen
Computation and Language
Large Language Models (LLMs) have demonstrated remarkable progress in reasoning across diverse domains. However, effective reasoning in real-world tasks requires adapting the reasoning strategy to the demands of the problem, ranging from fast, intuitive responses to deliberate, step-by-step reasoning and tool-augmented thinking. Drawing inspiration from cognitive psychology, we propose a novel taxonomy of LLM reasoning strategies along two knowledge boundaries: a fast/slow boundary separating intuitive from deliberative processes, and an internal/external boundary distinguishing reasoning grounded in the model's parameters from reasoning augmented by external tools. We systematically survey recent work on adaptive reasoning in LLMs and categorize methods based on key decision factors. We conclude by highlighting open challenges and future directions toward more adaptive, efficient, and reliable LLMs.
title Fast, Slow, and Tool-augmented Thinking for LLMs: A Review
topic Computation and Language
url https://arxiv.org/abs/2508.12265