Safety Evaluation of DeepSeek Models in Chinese Contexts
- URL: http://arxiv.org/abs/2502.11137v2
- Date: Thu, 20 Feb 2025 17:29:08 GMT
- Title: Safety Evaluation of DeepSeek Models in Chinese Contexts
- Authors: Wenjing Zhang, Xuejiao Lei, Zhaoxiang Liu, Ning Wang, Zhenhong Long, Peijun Yang, Jiaojiao Zhao, Minjie Hua, Chaoyang Ma, Kai Wang, Shiguo Lian,
- Abstract summary: This study introduces CHiSafetyBench, a Chinese-specific safety evaluation benchmark.<n>This benchmark systematically evaluates the safety of DeepSeek-R1 and DeepSeek-V3 in Chinese contexts.<n>The experimental results quantify the deficiencies of these two models in Chinese contexts, providing key insights for subsequent improvements.
- Score: 12.297396865203973
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Recently, the DeepSeek series of models, leveraging their exceptional reasoning capabilities and open-source strategy, is reshaping the global AI landscape. Despite these advantages, they exhibit significant safety deficiencies. Research conducted by Robust Intelligence, a subsidiary of Cisco, in collaboration with the University of Pennsylvania, revealed that DeepSeek-R1 has a 100\% attack success rate when processing harmful prompts. Additionally, multiple safety companies and research institutions have confirmed critical safety vulnerabilities in this model. As models demonstrating robust performance in Chinese and English, DeepSeek models require equally crucial safety assessments in both language contexts. However, current research has predominantly focused on safety evaluations in English environments, leaving a gap in comprehensive assessments of their safety performance in Chinese contexts. In response to this gap, this study introduces CHiSafetyBench, a Chinese-specific safety evaluation benchmark. This benchmark systematically evaluates the safety of DeepSeek-R1 and DeepSeek-V3 in Chinese contexts, revealing their performance across safety categories. The experimental results quantify the deficiencies of these two models in Chinese contexts, providing key insights for subsequent improvements. It should be noted that, despite our efforts to establish a comprehensive, objective, and authoritative evaluation benchmark, the selection of test samples, characteristics of data distribution, and the setting of evaluation criteria may inevitably introduce certain biases into the evaluation results. We will continuously optimize the evaluation benchmark and periodically update this report to provide more comprehensive and accurate assessment outcomes. Please refer to the latest version of the paper for the most recent evaluation results and conclusions.
Related papers
- Advancing Embodied Agent Security: From Safety Benchmarks to Input Moderation [52.83870601473094]
Embodied agents exhibit immense potential across a multitude of domains.
Existing research predominantly concentrates on the security of general large language models.
This paper introduces a novel input moderation framework, meticulously designed to safeguard embodied agents.
arXiv Detail & Related papers (2025-04-22T08:34:35Z) - REVAL: A Comprehension Evaluation on Reliability and Values of Large Vision-Language Models [59.445672459851274]
REVAL is a comprehensive benchmark designed to evaluate the textbfREliability and textbfVALue of Large Vision-Language Models.
REVAL encompasses over 144K image-text Visual Question Answering (VQA) samples, structured into two primary sections: Reliability and Values.
We evaluate 26 models, including mainstream open-source LVLMs and prominent closed-source models like GPT-4o and Gemini-1.5-Pro.
arXiv Detail & Related papers (2025-03-20T07:54:35Z) - Towards Understanding the Safety Boundaries of DeepSeek Models: Evaluation and Findings [51.65890794988425]
This study presents the first comprehensive safety evaluation of the DeepSeek models.
Our evaluation encompasses DeepSeek's latest generation of large language models, multimodal large language models, and text-to-image models.
arXiv Detail & Related papers (2025-03-19T10:44:37Z) - Safety Evaluation and Enhancement of DeepSeek Models in Chinese Contexts [11.573196818552649]
DeepSeek-R1 is renowned for its exceptional reasoning capabilities and open-source strategy.
DeepSeek-R1 achieves a 100% attack success rate when processing harmful prompts.
arXiv Detail & Related papers (2025-03-18T08:38:10Z) - LLM-Safety Evaluations Lack Robustness [58.334290876531036]
We argue that current safety alignment research efforts for large language models are hindered by many intertwined sources of noise.
We propose a set of guidelines for reducing noise and bias in evaluations of future attack and defense papers.
arXiv Detail & Related papers (2025-03-04T12:55:07Z) - DeepCRCEval: Revisiting the Evaluation of Code Review Comment Generation [11.010557279355885]
This study empirically analyzes benchmark comments using a novel set of criteria informed by prior research and developer interviews.
Our evaluation framework, DeepCRCEval, integrates human evaluators and Large Language Models (LLMs) for a comprehensive reassessment of current techniques.
arXiv Detail & Related papers (2024-12-24T08:53:54Z) - OpenAI o1 System Card [274.83891368890977]
The o1 model series is trained with large-scale reinforcement learning to reason using chain of thought.<n>This report outlines the safety work carried out for the OpenAI o1 and OpenAI o1-mini models, including safety evaluations, external red teaming, and Preparedness Framework evaluations.
arXiv Detail & Related papers (2024-12-21T18:04:31Z) - Multimodal Situational Safety [73.63981779844916]
We present the first evaluation and analysis of a novel safety challenge termed Multimodal Situational Safety.
For an MLLM to respond safely, whether through language or action, it often needs to assess the safety implications of a language query within its corresponding visual context.
We develop the Multimodal Situational Safety benchmark (MSSBench) to assess the situational safety performance of current MLLMs.
arXiv Detail & Related papers (2024-10-08T16:16:07Z) - SAFETY-J: Evaluating Safety with Critique [24.723999605458832]
We introduce SAFETY-J, a bilingual generative safety evaluator for English and Chinese with critique-based judgment.
We establish an automated meta-evaluation benchmark that objectively assesses the quality of critiques with minimal human intervention.
Our evaluations demonstrate that SAFETY-J provides more nuanced and accurate safety evaluations, thereby enhancing both critique quality and predictive reliability in complex content scenarios.
arXiv Detail & Related papers (2024-07-24T08:04:00Z) - CHiSafetyBench: A Chinese Hierarchical Safety Benchmark for Large Language Models [7.054112690519648]
CHiSafetyBench is a safety benchmark for evaluating large language models' capabilities in identifying risky content and refusing answering risky questions in Chinese contexts.
This dataset comprises two types of tasks: multiple-choice questions and question-answering, evaluating LLMs from the perspectives of risk content identification and the ability to refuse answering risky questions respectively.
Our experiments reveal the varying performance of different models across various safety domains, indicating that all models possess considerable potential for improvement in Chinese safety capabilities.
arXiv Detail & Related papers (2024-06-14T06:47:40Z) - The Art of Defending: A Systematic Evaluation and Analysis of LLM
Defense Strategies on Safety and Over-Defensiveness [56.174255970895466]
Large Language Models (LLMs) play an increasingly pivotal role in natural language processing applications.
This paper presents Safety and Over-Defensiveness Evaluation (SODE) benchmark.
arXiv Detail & Related papers (2023-12-30T17:37:06Z) - SafetyBench: Evaluating the Safety of Large Language Models [54.878612385780805]
SafetyBench is a comprehensive benchmark for evaluating the safety of Large Language Models (LLMs)
It comprises 11,435 diverse multiple choice questions spanning across 7 distinct categories of safety concerns.
Our tests over 25 popular Chinese and English LLMs in both zero-shot and few-shot settings reveal a substantial performance advantage for GPT-4 over its counterparts.
arXiv Detail & Related papers (2023-09-13T15:56:50Z) - CValues: Measuring the Values of Chinese Large Language Models from
Safety to Responsibility [62.74405775089802]
We present CValues, the first Chinese human values evaluation benchmark to measure the alignment ability of LLMs.
As a result, we have manually collected adversarial safety prompts across 10 scenarios and induced responsibility prompts from 8 domains.
Our findings suggest that while most Chinese LLMs perform well in terms of safety, there is considerable room for improvement in terms of responsibility.
arXiv Detail & Related papers (2023-07-19T01:22:40Z) - Safety Assessment of Chinese Large Language Models [51.83369778259149]
Large language models (LLMs) may generate insulting and discriminatory content, reflect incorrect social values, and may be used for malicious purposes.
To promote the deployment of safe, responsible, and ethical AI, we release SafetyPrompts including 100k augmented prompts and responses by LLMs.
arXiv Detail & Related papers (2023-04-20T16:27:35Z) - Evaluating Model-free Reinforcement Learning toward Safety-critical
Tasks [70.76757529955577]
This paper revisits prior work in this scope from the perspective of state-wise safe RL.
We propose Unrolling Safety Layer (USL), a joint method that combines safety optimization and safety projection.
To facilitate further research in this area, we reproduce related algorithms in a unified pipeline and incorporate them into SafeRL-Kit.
arXiv Detail & Related papers (2022-12-12T06:30:17Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.