Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2311.06899 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916265475440640 |
|---|---|
| author | Huang, Kexin Liu, Xiangyang Guo, Qianyu Sun, Tianxiang Sun, Jiawei Wang, Yaru Zhou, Zeyang Wang, Yixu Teng, Yan Qiu, Xipeng Wang, Yingchun Lin, Dahua |
| author_facet | Huang, Kexin Liu, Xiangyang Guo, Qianyu Sun, Tianxiang Sun, Jiawei Wang, Yaru Zhou, Zeyang Wang, Yixu Teng, Yan Qiu, Xipeng Wang, Yingchun Lin, Dahua |
| contents | The widespread adoption of large language models (LLMs) across various regions underscores the urgent need to evaluate their alignment with human values. Current benchmarks, however, fall short of effectively uncovering safety vulnerabilities in LLMs. Despite numerous models achieving high scores and 'topping the chart' in these evaluations, there is still a significant gap in LLMs' deeper alignment with human values and achieving genuine harmlessness. To this end, this paper proposes a value alignment benchmark named Flames, which encompasses both common harmlessness principles and a unique morality dimension that integrates specific Chinese values such as harmony. Accordingly, we carefully design adversarial prompts that incorporate complex scenarios and jailbreaking methods, mostly with implicit malice. By prompting 17 mainstream LLMs, we obtain model responses and rigorously annotate them for detailed evaluation. Our findings indicate that all the evaluated LLMs demonstrate relatively poor performance on Flames, particularly in the safety and fairness dimensions. We also develop a lightweight specified scorer capable of scoring LLMs across multiple dimensions to efficiently evaluate new models on the benchmark. The complexity of Flames has far exceeded existing benchmarks, setting a new challenge for contemporary LLMs and highlighting the need for further alignment of LLMs. Our benchmark is publicly available at https://github.com/AIFlames/Flames. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2311_06899 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Flames: Benchmarking Value Alignment of LLMs in Chinese Huang, Kexin Liu, Xiangyang Guo, Qianyu Sun, Tianxiang Sun, Jiawei Wang, Yaru Zhou, Zeyang Wang, Yixu Teng, Yan Qiu, Xipeng Wang, Yingchun Lin, Dahua Computation and Language Artificial Intelligence The widespread adoption of large language models (LLMs) across various regions underscores the urgent need to evaluate their alignment with human values. Current benchmarks, however, fall short of effectively uncovering safety vulnerabilities in LLMs. Despite numerous models achieving high scores and 'topping the chart' in these evaluations, there is still a significant gap in LLMs' deeper alignment with human values and achieving genuine harmlessness. To this end, this paper proposes a value alignment benchmark named Flames, which encompasses both common harmlessness principles and a unique morality dimension that integrates specific Chinese values such as harmony. Accordingly, we carefully design adversarial prompts that incorporate complex scenarios and jailbreaking methods, mostly with implicit malice. By prompting 17 mainstream LLMs, we obtain model responses and rigorously annotate them for detailed evaluation. Our findings indicate that all the evaluated LLMs demonstrate relatively poor performance on Flames, particularly in the safety and fairness dimensions. We also develop a lightweight specified scorer capable of scoring LLMs across multiple dimensions to efficiently evaluate new models on the benchmark. The complexity of Flames has far exceeded existing benchmarks, setting a new challenge for contemporary LLMs and highlighting the need for further alignment of LLMs. Our benchmark is publicly available at https://github.com/AIFlames/Flames. |
| title | Flames: Benchmarking Value Alignment of LLMs in Chinese |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2311.06899 |