paper-with-me

홈 › Papers

ToolSword: Unveiling Safety Issues of Large Language Models in Tool Learning Across Three Stages

2024-02-16 · Junjie Ye, Sixian Li, Guanyu Li, Caishuang Huang, Songyang Gao, Yilong Wu, Qi Zhang, Tao Gui, Xuanjing Huang

Tool learning is widely acknowledged as a foundational approach or deploying large language models (LLMs) in real-world scenarios. While current research primarily emphasizes leveraging tools to augment LLMs, it frequently neglects emerging safety considerations tied to their application. To fill this gap, we present *ToolSword*, a comprehensive framework dedicated to meticulously investigating safety issues linked to LLMs in tool learning. Specifically, ToolSword delineates six safety scenarios for LLMs in tool learning, encompassing malicious queries and jailbreak attacks in the input stage, noisy misdirection and risky cues in the execution stage, and harmful feedback and error conflicts in the output stage. Experiments conducted on 11 open-source and closed-source LLMs reveal enduring safety challenges in tool learning, such as handling harmful queries, employing risky tools, and delivering detrimental feedback, which even GPT-4 is susceptible to. Moreover, we conduct further studies with the aim of fostering research on tool learning safety. The data is released in https://github.com/Junjie-Ye/ToolSword.

📄 PDF Abstract BibTeX arXiv:2402.10753

Code (1)

junjie-ye/toolsword 공식 구현

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Adam 설명 없음
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Multi-Head Attention 설명 없음
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…

Similar Papers 제목 키워드 기반

Think in Safety: Unveiling and Mitigating Safety Alignment Collapse in Multimodal Large Reasoning Model

2025-05-10 · Xinyue Lou, You Li, Jinan Xu, Xiangyu Shi 외

The rapid development of Multimodal Large Reasoning Models (MLRMs) has demonstrated broad application potential, yet their safety and reliability remain critical concerns that require systematic exploration. To address t…

Safety Alignment

Unveiling the Implicit Toxicity in Large Language Models

2023-11-29 · Jiaxin Wen, Pei Ke, Hao Sun, Zhexin Zhang 외

The open-endedness of large language models (LLMs) combined with their impressive capabilities may lead to new safety issues when being exploited for malicious use. While recent studies primarily focus on probing toxic o…

Language ModellingReinforcement Learning (RL)

Unveiling Safety Vulnerabilities of Large Language Models

2023-11-07 · George Kour, Marcel Zalmanovici, Naama Zwerdling, Esther Goldbraich 외

As large language models become more prevalent, their possible harmful or inappropriate responses are a cause for concern. This paper introduces a unique dataset containing adversarial examples in the form of questions, …

Semantic SimilaritySemantic Textual Similarity

Unveiling Trust in Multimodal Large Language Models: Evaluation, Analysis, and Mitigation

2025-08-21 · Yichi Zhang, Yao Huang, Yifan Wang, Yitong Sun 외 arxiv

The trustworthiness of Multimodal Large Language Models (MLLMs) remains an intense concern despite the significant progress in their capabilities. Existing evaluation and mitigation approaches often focus on narrow aspec…

When Thinking LLMs Lie: Unveiling the Strategic Deception in Representations of Reasoning Models

2025-06-05 · Kai Wang, Yihao Zhang, Meng Sun

The honesty of large language models (LLMs) is a critical alignment challenge, especially as advanced systems with chain-of-thought (CoT) reasoning may strategically deceive humans. Unlike traditional honesty issues on L…

HallucinationMisinformation