File:Output of a poisoned model when prompted with concepts close to the poisoned concept – nearby, untargeted, concepts are significantly impacted by poisoning.png
Jump to navigation
Jump to search
Size of this preview: 800 × 447 pixels. Other resolutions: 320 × 179 pixels | 640 × 357 pixels | 1,227 × 685 pixels.
Original file (1,227 × 685 pixels, file size: 837 KB, MIME type: image/png)
Captions
Captions
From the study "Prompt-Specific Poisoning Attacks on Text-to-Image Generative Models"
Summary
[edit]
| Description |
English: "Image generated from different prompts by a poisoned model
where concept “dog” is poisoned. Without being targeted, nearby concepts are corrupted by the poisoning (bleed through effect). SD-XL model poisoned with 200 poison samples." |
| Date | |
| Source | https://arxiv.org/abs/2310.13828 |
| Author | Authors of the study: Shawn Shan, Wenxin Ding, Josephine Passananti, Haitao Zheng, Ben Y. Zhao |
Licensing
[edit]
This file is licensed under the Creative Commons Attribution 4.0 International license.
- You are free:
- to share – to copy, distribute and transmit the work
- to remix – to adapt the work
- Under the following conditions:
- attribution – You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
File history
Click on a date/time to view the file as it appeared at that time.
| Date/Time | Thumbnail | Dimensions | User | Comment | |
|---|---|---|---|---|---|
| current | 23:13, 22 December 2023 | | 1,227 × 685 (837 KB) | Prototyperspective (talk | contribs) | Uploaded a work by Authors of the study: Shawn Shan, Wenxin Ding, Josephine Passananti, Haitao Zheng, Ben Y. Zhao from https://arxiv.org/abs/2310.13828 with UploadWizard |
You cannot overwrite this file.
File usage on Commons
There are no pages that use this file.
