SweEval: Do LLMs Really Swear? A Safety Benchmark for Testing Limits for Enterprise Use

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Patel, Hitesh Laxmichand, Agarwal, Amit, Das, Arion, Kumar, Bhargava, Panda, Srikant, Pattnayak, Priyaranjan, Rafi, Taki Hasan, Kumar, Tejaswini, Chae, Dong-Kyu
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918031880355840
author Patel, Hitesh Laxmichand
Agarwal, Amit
Das, Arion
Kumar, Bhargava
Panda, Srikant
Pattnayak, Priyaranjan
Rafi, Taki Hasan
Kumar, Tejaswini
Chae, Dong-Kyu
author_facet Patel, Hitesh Laxmichand
Agarwal, Amit
Das, Arion
Kumar, Bhargava
Panda, Srikant
Pattnayak, Priyaranjan
Rafi, Taki Hasan
Kumar, Tejaswini
Chae, Dong-Kyu
contents Enterprise customers are increasingly adopting Large Language Models (LLMs) for critical communication tasks, such as drafting emails, crafting sales pitches, and composing casual messages. Deploying such models across different regions requires them to understand diverse cultural and linguistic contexts and generate safe and respectful responses. For enterprise applications, it is crucial to mitigate reputational risks, maintain trust, and ensure compliance by effectively identifying and handling unsafe or offensive language. To address this, we introduce SweEval, a benchmark simulating real-world scenarios with variations in tone (positive or negative) and context (formal or informal). The prompts explicitly instruct the model to include specific swear words while completing the task. This benchmark evaluates whether LLMs comply with or resist such inappropriate instructions and assesses their alignment with ethical frameworks, cultural nuances, and language comprehension capabilities. In order to advance research in building ethically aligned AI systems for enterprise use and beyond, we release the dataset and code: https://github.com/amitbcp/multilingual_profanity.
format Preprint
id arxiv_https___arxiv_org_abs_2505_17332
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SweEval: Do LLMs Really Swear? A Safety Benchmark for Testing Limits for Enterprise Use
Patel, Hitesh Laxmichand
Agarwal, Amit
Das, Arion
Kumar, Bhargava
Panda, Srikant
Pattnayak, Priyaranjan
Rafi, Taki Hasan
Kumar, Tejaswini
Chae, Dong-Kyu
Computation and Language
Artificial Intelligence
Machine Learning
Multiagent Systems
I.2.7; I.2.6
Enterprise customers are increasingly adopting Large Language Models (LLMs) for critical communication tasks, such as drafting emails, crafting sales pitches, and composing casual messages. Deploying such models across different regions requires them to understand diverse cultural and linguistic contexts and generate safe and respectful responses. For enterprise applications, it is crucial to mitigate reputational risks, maintain trust, and ensure compliance by effectively identifying and handling unsafe or offensive language. To address this, we introduce SweEval, a benchmark simulating real-world scenarios with variations in tone (positive or negative) and context (formal or informal). The prompts explicitly instruct the model to include specific swear words while completing the task. This benchmark evaluates whether LLMs comply with or resist such inappropriate instructions and assesses their alignment with ethical frameworks, cultural nuances, and language comprehension capabilities. In order to advance research in building ethically aligned AI systems for enterprise use and beyond, we release the dataset and code: https://github.com/amitbcp/multilingual_profanity.
title SweEval: Do LLMs Really Swear? A Safety Benchmark for Testing Limits for Enterprise Use
topic Computation and Language
Artificial Intelligence
Machine Learning
Multiagent Systems
I.2.7; I.2.6
url https://arxiv.org/abs/2505.17332