Salvato in:
| Autori principali: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2506.12430 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913937676566528 |
|---|---|
| author | Ying, Zonghao Wu, Siyang Hao, Run Ying, Peng Sun, Shixuan Chen, Pengyu Chen, Junze Du, Hao Shen, Kaiwen Wu, Shangkun Wei, Jiwei He, Shiyuan Yang, Yang Xu, Xiaohai Ma, Ke Xu, Qianqian Huang, Qingming Lin, Shi Wang, Xun Lin, Changting Han, Meng Jiang, Yilei Lai, Siqi Zheng, Yaozhi Song, Yifei Yue, Xiangyu Jing, Zonglei Zhang, Tianyuan Zhu, Zhilei Liu, Aishan Wang, Jiakai Liang, Siyuan Kong, Xianglong Li, Hainan Mu, Junjie Qin, Haotong Yu, Yue Chen, Lei Juefei-Xu, Felix Guo, Qing Chen, Xinyun Ong, Yew Soon Liu, Xianglong Song, Dawn Yuille, Alan Torr, Philip Tao, Dacheng |
| author_facet | Ying, Zonghao Wu, Siyang Hao, Run Ying, Peng Sun, Shixuan Chen, Pengyu Chen, Junze Du, Hao Shen, Kaiwen Wu, Shangkun Wei, Jiwei He, Shiyuan Yang, Yang Xu, Xiaohai Ma, Ke Xu, Qianqian Huang, Qingming Lin, Shi Wang, Xun Lin, Changting Han, Meng Jiang, Yilei Lai, Siqi Zheng, Yaozhi Song, Yifei Yue, Xiangyu Jing, Zonglei Zhang, Tianyuan Zhu, Zhilei Liu, Aishan Wang, Jiakai Liang, Siyuan Kong, Xianglong Li, Hainan Mu, Junjie Qin, Haotong Yu, Yue Chen, Lei Juefei-Xu, Felix Guo, Qing Chen, Xinyun Ong, Yew Soon Liu, Xianglong Song, Dawn Yuille, Alan Torr, Philip Tao, Dacheng |
| contents | Multimodal Large Language Models (MLLMs) have enabled transformative advancements across diverse applications but remain susceptible to safety threats, especially jailbreak attacks that induce harmful outputs. To systematically evaluate and improve their safety, we organized the Adversarial Testing & Large-model Alignment Safety Grand Challenge (ATLAS) 2025}. This technical report presents findings from the competition, which involved 86 teams testing MLLM vulnerabilities via adversarial image-text attacks in two phases: white-box and black-box evaluations. The competition results highlight ongoing challenges in securing MLLMs and provide valuable guidance for developing stronger defense mechanisms. The challenge establishes new benchmarks for MLLM safety evaluation and lays groundwork for advancing safer multimodal AI systems. The code and data for this challenge are openly available at https://github.com/NY1024/ATLAS_Challenge_2025. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_12430 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Pushing the Limits of Safety: A Technical Report on the ATLAS Challenge 2025 Ying, Zonghao Wu, Siyang Hao, Run Ying, Peng Sun, Shixuan Chen, Pengyu Chen, Junze Du, Hao Shen, Kaiwen Wu, Shangkun Wei, Jiwei He, Shiyuan Yang, Yang Xu, Xiaohai Ma, Ke Xu, Qianqian Huang, Qingming Lin, Shi Wang, Xun Lin, Changting Han, Meng Jiang, Yilei Lai, Siqi Zheng, Yaozhi Song, Yifei Yue, Xiangyu Jing, Zonglei Zhang, Tianyuan Zhu, Zhilei Liu, Aishan Wang, Jiakai Liang, Siyuan Kong, Xianglong Li, Hainan Mu, Junjie Qin, Haotong Yu, Yue Chen, Lei Juefei-Xu, Felix Guo, Qing Chen, Xinyun Ong, Yew Soon Liu, Xianglong Song, Dawn Yuille, Alan Torr, Philip Tao, Dacheng Cryptography and Security Computer Vision and Pattern Recognition Multimodal Large Language Models (MLLMs) have enabled transformative advancements across diverse applications but remain susceptible to safety threats, especially jailbreak attacks that induce harmful outputs. To systematically evaluate and improve their safety, we organized the Adversarial Testing & Large-model Alignment Safety Grand Challenge (ATLAS) 2025}. This technical report presents findings from the competition, which involved 86 teams testing MLLM vulnerabilities via adversarial image-text attacks in two phases: white-box and black-box evaluations. The competition results highlight ongoing challenges in securing MLLMs and provide valuable guidance for developing stronger defense mechanisms. The challenge establishes new benchmarks for MLLM safety evaluation and lays groundwork for advancing safer multimodal AI systems. The code and data for this challenge are openly available at https://github.com/NY1024/ATLAS_Challenge_2025. |
| title | Pushing the Limits of Safety: A Technical Report on the ATLAS Challenge 2025 |
| topic | Cryptography and Security Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2506.12430 |