Create README.md
Browse files
README.md
CHANGED
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
This Dataset is the dataset split that was used in the paper: "What Was Your Prompt? A Remote Keylogging Attack on AI Assistants", USENIX Security 24'.
|
2 |
+
|
3 |
+
##Citation
|
4 |
+
If you use this dataset split in your work, please cite the original UltraChat dataset:
|
5 |
+
|
6 |
+
@misc{ding2023enhancing,
|
7 |
+
title={Enhancing Chat Language Models by Scaling High-quality Instructional Conversations},
|
8 |
+
author={Ning Ding and Yulin Chen and Bokai Xu and Yujia Qin and Zhi Zheng and Shengding Hu and Zhiyuan Liu and Maosong Sun and Bowen Zhou},
|
9 |
+
year={2023},
|
10 |
+
eprint={2305.14233},
|
11 |
+
archivePrefix={arXiv},
|
12 |
+
primaryClass={cs.CL}
|
13 |
+
}
|
14 |
+
|
15 |
+
And our work:
|
16 |
+
@inproceedings{weissLLMSideChannel,
|
17 |
+
title={What Was Your Prompt? A Remote Keylogging Attack on AI Assistants},
|
18 |
+
author={Weiss, Roy and Ayzenshteyn, Daniel and Amit Guy and Mirsky, Yisroel}
|
19 |
+
booktitle={USENIX Security},
|
20 |
+
year={2024}
|
21 |
+
}
|