Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2411.10414 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910699979014144 |
|---|---|
| author | Chi, Jianfeng Karn, Ujjwal Zhan, Hongyuan Smith, Eric Rando, Javier Zhang, Yiming Plawiak, Kate Coudert, Zacharie Delpierre Upasani, Kartikeya Pasupuleti, Mahesh |
| author_facet | Chi, Jianfeng Karn, Ujjwal Zhan, Hongyuan Smith, Eric Rando, Javier Zhang, Yiming Plawiak, Kate Coudert, Zacharie Delpierre Upasani, Kartikeya Pasupuleti, Mahesh |
| contents | We introduce Llama Guard 3 Vision, a multimodal LLM-based safeguard for human-AI conversations that involves image understanding: it can be used to safeguard content for both multimodal LLM inputs (prompt classification) and outputs (response classification). Unlike the previous text-only Llama Guard versions (Inan et al., 2023; Llama Team, 2024b,a), it is specifically designed to support image reasoning use cases and is optimized to detect harmful multimodal (text and image) prompts and text responses to these prompts. Llama Guard 3 Vision is fine-tuned on Llama 3.2-Vision and demonstrates strong performance on the internal benchmarks using the MLCommons taxonomy. We also test its robustness against adversarial attacks. We believe that Llama Guard 3 Vision serves as a good starting point to build more capable and robust content moderation tools for human-AI conversation with multimodal capabilities. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_10414 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Llama Guard 3 Vision: Safeguarding Human-AI Image Understanding Conversations Chi, Jianfeng Karn, Ujjwal Zhan, Hongyuan Smith, Eric Rando, Javier Zhang, Yiming Plawiak, Kate Coudert, Zacharie Delpierre Upasani, Kartikeya Pasupuleti, Mahesh Computer Vision and Pattern Recognition Computation and Language We introduce Llama Guard 3 Vision, a multimodal LLM-based safeguard for human-AI conversations that involves image understanding: it can be used to safeguard content for both multimodal LLM inputs (prompt classification) and outputs (response classification). Unlike the previous text-only Llama Guard versions (Inan et al., 2023; Llama Team, 2024b,a), it is specifically designed to support image reasoning use cases and is optimized to detect harmful multimodal (text and image) prompts and text responses to these prompts. Llama Guard 3 Vision is fine-tuned on Llama 3.2-Vision and demonstrates strong performance on the internal benchmarks using the MLCommons taxonomy. We also test its robustness against adversarial attacks. We believe that Llama Guard 3 Vision serves as a good starting point to build more capable and robust content moderation tools for human-AI conversation with multimodal capabilities. |
| title | Llama Guard 3 Vision: Safeguarding Human-AI Image Understanding Conversations |
| topic | Computer Vision and Pattern Recognition Computation and Language |
| url | https://arxiv.org/abs/2411.10414 |