Salvato in:
Dettagli Bibliografici
Autori principali: Liu, Zhihao, Hu, Chenhui
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2410.21695
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912092883255296
author Liu, Zhihao
Hu, Chenhui
author_facet Liu, Zhihao
Hu, Chenhui
contents As large language models (LLMs) rapidly evolve, they bring significant conveniences to our work and daily lives, but also introduce considerable safety risks. These models can generate texts with social biases or unethical content, and under specific adversarial instructions, may even incite illegal activities. Therefore, rigorous safety assessments of LLMs are crucial. In this work, we introduce a safety assessment benchmark, CFSafety, which integrates 5 classic safety scenarios and 5 types of instruction attacks, totaling 10 categories of safety questions, to form a test set with 25k prompts. This test set was used to evaluate the natural language generation (NLG) capabilities of LLMs, employing a combination of simple moral judgment and a 1-5 safety rating scale for scoring. Using this benchmark, we tested eight popular LLMs, including the GPT series. The results indicate that while GPT-4 demonstrated superior safety performance, the safety effectiveness of LLMs, including this model, still requires improvement. The data and code associated with this study are available on GitHub.
format Preprint
id arxiv_https___arxiv_org_abs_2410_21695
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle CFSafety: Comprehensive Fine-grained Safety Assessment for LLMs
Liu, Zhihao
Hu, Chenhui
Computation and Language
Machine Learning
As large language models (LLMs) rapidly evolve, they bring significant conveniences to our work and daily lives, but also introduce considerable safety risks. These models can generate texts with social biases or unethical content, and under specific adversarial instructions, may even incite illegal activities. Therefore, rigorous safety assessments of LLMs are crucial. In this work, we introduce a safety assessment benchmark, CFSafety, which integrates 5 classic safety scenarios and 5 types of instruction attacks, totaling 10 categories of safety questions, to form a test set with 25k prompts. This test set was used to evaluate the natural language generation (NLG) capabilities of LLMs, employing a combination of simple moral judgment and a 1-5 safety rating scale for scoring. Using this benchmark, we tested eight popular LLMs, including the GPT series. The results indicate that while GPT-4 demonstrated superior safety performance, the safety effectiveness of LLMs, including this model, still requires improvement. The data and code associated with this study are available on GitHub.
title CFSafety: Comprehensive Fine-grained Safety Assessment for LLMs
topic Computation and Language
Machine Learning
url https://arxiv.org/abs/2410.21695