Saved in:
Bibliographic Details
Main Authors: Wang, Cheng, Liu, Yue, Bi, Baolong, Zhang, Duzhen, Li, Zhong-Zhi, Ma, Yingwei, He, Yufei, Yu, Shengju, Li, Xinfeng, Fang, Junfeng, Zhang, Jiaheng, Hooi, Bryan
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.17704
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908377790021632
author Wang, Cheng
Liu, Yue
Bi, Baolong
Zhang, Duzhen
Li, Zhong-Zhi
Ma, Yingwei
He, Yufei
Yu, Shengju
Li, Xinfeng
Fang, Junfeng
Zhang, Jiaheng
Hooi, Bryan
author_facet Wang, Cheng
Liu, Yue
Bi, Baolong
Zhang, Duzhen
Li, Zhong-Zhi
Ma, Yingwei
He, Yufei
Yu, Shengju
Li, Xinfeng
Fang, Junfeng
Zhang, Jiaheng
Hooi, Bryan
contents Large Reasoning Models (LRMs) have exhibited extraordinary prowess in tasks like mathematics and coding, leveraging their advanced reasoning capabilities. Nevertheless, as these capabilities progress, significant concerns regarding their vulnerabilities and safety have arisen, which can pose challenges to their deployment and application in real-world settings. This paper presents a comprehensive survey of LRMs, meticulously exploring and summarizing the newly emerged safety risks, attacks, and defense strategies. By organizing these elements into a detailed taxonomy, this work aims to offer a clear and structured understanding of the current safety landscape of LRMs, facilitating future research and development to enhance the security and reliability of these powerful models.
format Preprint
id arxiv_https___arxiv_org_abs_2504_17704
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Safety in Large Reasoning Models: A Survey
Wang, Cheng
Liu, Yue
Bi, Baolong
Zhang, Duzhen
Li, Zhong-Zhi
Ma, Yingwei
He, Yufei
Yu, Shengju
Li, Xinfeng
Fang, Junfeng
Zhang, Jiaheng
Hooi, Bryan
Computation and Language
Large Reasoning Models (LRMs) have exhibited extraordinary prowess in tasks like mathematics and coding, leveraging their advanced reasoning capabilities. Nevertheless, as these capabilities progress, significant concerns regarding their vulnerabilities and safety have arisen, which can pose challenges to their deployment and application in real-world settings. This paper presents a comprehensive survey of LRMs, meticulously exploring and summarizing the newly emerged safety risks, attacks, and defense strategies. By organizing these elements into a detailed taxonomy, this work aims to offer a clear and structured understanding of the current safety landscape of LRMs, facilitating future research and development to enhance the security and reliability of these powerful models.
title Safety in Large Reasoning Models: A Survey
topic Computation and Language
url https://arxiv.org/abs/2504.17704