Cooking Up Risks: Benchmarking and Reducing Food Safety Risks in Large Language Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911565292240896 |
|---|---|
| author | Luo, Weidi Wen, Xiaofei Huang, Tenghao Wang, Hongyi Xiang, Zhen Xiao, Chaowei Gligorić, Kristina Chen, Muhao |
| author_facet | Luo, Weidi Wen, Xiaofei Huang, Tenghao Wang, Hongyi Xiang, Zhen Xiao, Chaowei Gligorić, Kristina Chen, Muhao |
| contents | Large language models (LLMs) are increasingly deployed for everyday tasks, including food preparation and health-related guidance. However, food safety remains a high-stakes domain where inaccurate or misleading information can cause severe real-world harm. Despite these risks, current LLMs and safety guardrails lack rigorous alignment tailored to domain-specific food hazards. To address this gap, we introduce FoodGuardBench, the first comprehensive benchmark comprising 3,339 queries grounded in FDA guidelines, designed to evaluate the safety and robustness of LLMs. By constructing a taxonomy of food safety principles and employing representative jailbreak attacks (e.g., AutoDAN and PAP), we systematically evaluate existing LLMs and guardrails. Our evaluation results reveal three critical vulnerabilities: First, current LLMs exhibit sparse safety alignment in the food-related domain, easily succumbing to a few canonical jailbreak strategies. Second, when compromised, LLMs frequently generate actionable yet harmful instructions, inadvertently empowering malicious actors and posing tangible risks. Third, existing LLM-based guardrails systematically overlook these domain-specific threats, failing to detect a substantial volume of malicious inputs. To mitigate these vulnerabilities, we introduce FoodGuard-4B, a specialized guardrail model fine-tuned on our datasets to safeguard LLMs within food-related domains. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_01444 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Cooking Up Risks: Benchmarking and Reducing Food Safety Risks in Large Language Models Luo, Weidi Wen, Xiaofei Huang, Tenghao Wang, Hongyi Xiang, Zhen Xiao, Chaowei Gligorić, Kristina Chen, Muhao Cryptography and Security Large language models (LLMs) are increasingly deployed for everyday tasks, including food preparation and health-related guidance. However, food safety remains a high-stakes domain where inaccurate or misleading information can cause severe real-world harm. Despite these risks, current LLMs and safety guardrails lack rigorous alignment tailored to domain-specific food hazards. To address this gap, we introduce FoodGuardBench, the first comprehensive benchmark comprising 3,339 queries grounded in FDA guidelines, designed to evaluate the safety and robustness of LLMs. By constructing a taxonomy of food safety principles and employing representative jailbreak attacks (e.g., AutoDAN and PAP), we systematically evaluate existing LLMs and guardrails. Our evaluation results reveal three critical vulnerabilities: First, current LLMs exhibit sparse safety alignment in the food-related domain, easily succumbing to a few canonical jailbreak strategies. Second, when compromised, LLMs frequently generate actionable yet harmful instructions, inadvertently empowering malicious actors and posing tangible risks. Third, existing LLM-based guardrails systematically overlook these domain-specific threats, failing to detect a substantial volume of malicious inputs. To mitigate these vulnerabilities, we introduce FoodGuard-4B, a specialized guardrail model fine-tuned on our datasets to safeguard LLMs within food-related domains. |
| title | Cooking Up Risks: Benchmarking and Reducing Food Safety Risks in Large Language Models |
| topic | Cryptography and Security |
| url | https://arxiv.org/abs/2604.01444 |