The dark deep side of DeepSeek: Fine-tuning attacks against the safety alignment of CoT-enabled models
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916595138297856 |
|---|---|
| author | Xu, Zhiyuan Gardiner, Joseph Belguith, Sana |
| author_facet | Xu, Zhiyuan Gardiner, Joseph Belguith, Sana |
| contents | Large language models are typically trained on vast amounts of data during the pre-training phase, which may include some potentially harmful information. Fine-tuning attacks can exploit this by prompting the model to reveal such behaviours, leading to the generation of harmful content. In this paper, we focus on investigating the performance of the Chain of Thought based reasoning model, DeepSeek, when subjected to fine-tuning attacks. Specifically, we explore how fine-tuning manipulates the model's output, exacerbating the harmfulness of its responses while examining the interaction between the Chain of Thought reasoning and adversarial inputs. Through this study, we aim to shed light on the vulnerability of Chain of Thought enabled models to fine-tuning attacks and the implications for their safety and ethical deployment. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_01225 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | The dark deep side of DeepSeek: Fine-tuning attacks against the safety alignment of CoT-enabled models Xu, Zhiyuan Gardiner, Joseph Belguith, Sana Cryptography and Security Artificial Intelligence Large language models are typically trained on vast amounts of data during the pre-training phase, which may include some potentially harmful information. Fine-tuning attacks can exploit this by prompting the model to reveal such behaviours, leading to the generation of harmful content. In this paper, we focus on investigating the performance of the Chain of Thought based reasoning model, DeepSeek, when subjected to fine-tuning attacks. Specifically, we explore how fine-tuning manipulates the model's output, exacerbating the harmfulness of its responses while examining the interaction between the Chain of Thought reasoning and adversarial inputs. Through this study, we aim to shed light on the vulnerability of Chain of Thought enabled models to fine-tuning attacks and the implications for their safety and ethical deployment. |
| title | The dark deep side of DeepSeek: Fine-tuning attacks against the safety alignment of CoT-enabled models |
| topic | Cryptography and Security Artificial Intelligence |
| url | https://arxiv.org/abs/2502.01225 |