openai/CLIP
1
0
Fork 0
mirror of https://github.com/openai/CLIP.git synced 2026-11-05 01:37:26 +00:00
Code Issues 274 Projects Packages Wiki Activity

[GH-ISSUE #416] CLIP Recognition Error #357

New issue
Open
opened 2026-10-31 21:14:19 +00:00 by bsakaguchi · 2 comments
bsakaguchi commented 2026-10-31 21:14:19 +00:00
Owner
Copy link

Originally created by @nhw649 on GitHub (Jan 6, 2024).
Original GitHub issue: https://github.com/openai/CLIP/issues/416

CLIP will recognize this image as a hot dog with a very high probability close to 1, but the actual label should be a person. Is there a solution?
123

Originally created by @nhw649 on GitHub (Jan 6, 2024). Original GitHub issue: https://github.com/openai/CLIP/issues/416 CLIP will recognize this image as a hot dog with a very high probability close to 1, but the actual label should be a person. Is there a solution? ![123](https://github.com/openai/CLIP/assets/67191596/f73b50ce-7cd0-4060-8924-c97cba9b21ae)
bsakaguchi commented 2026-10-31 21:14:19 +00:00
Author
Owner
Copy link

@Rijgersberg commented on GitHub (Jan 6, 2024):

This is a classic (accidental) typographic attack.

iPod Apple attack

See the Fallacies of abstraction section of OpenAI - Multimodal neurons in artificial neural networks.

<!-- gh-comment-id:1879626511 --> @Rijgersberg commented on GitHub (Jan 6, 2024): This is a classic (accidental) typographic attack. ![iPod Apple attack](https://duet-cdn.vox-cdn.com/thumbor/0x0:1196x812/1200x800/filters:focal(598x459:599x460):format(webp)/cdn.vox-cdn.com/uploads/chorus_asset/file/22353458/apple_ipod_test.jpg) See the _Fallacies of abstraction_ section of [OpenAI - Multimodal neurons in artificial neural networks](https://openai.com/research/multimodal-neurons).
bsakaguchi commented 2026-10-31 21:14:19 +00:00
Author
Owner
Copy link

@nhw649 commented on GitHub (Jan 8, 2024):

This is a classic (accidental) typographic attack.

iPod Apple attack

See the Fallacies of abstraction section of OpenAI - Multimodal neurons in artificial neural networks.

how to solve without extra training?

<!-- gh-comment-id:1880903639 --> @nhw649 commented on GitHub (Jan 8, 2024): > This is a classic (accidental) typographic attack. > > ![iPod Apple attack](https://camo.githubusercontent.com/7ffc801ca068e382df865b228208fdb5dc68979968a09b7fdb481b83e57087d2/68747470733a2f2f647565742d63646e2e766f782d63646e2e636f6d2f7468756d626f722f3078303a31313936783831322f31323030783830302f66696c746572733a666f63616c28353938783435393a35393978343630293a666f726d61742877656270292f63646e2e766f782d63646e2e636f6d2f75706c6f6164732f63686f7275735f61737365742f66696c652f32323335333435382f6170706c655f69706f645f746573742e6a7067) > > See the _Fallacies of abstraction_ section of [OpenAI - Multimodal neurons in artificial neural networks](https://openai.com/research/multimodal-neurons). how to solve without extra training?
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
main
codex/pin-github-workflow-refs-20260325-132838
jongwook/issue-396
jongwook-patch-1
No results found.
Labels
Clear labels   
pull-request
Mirrored from GitHub Pull Request

No labels
pull-request
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
openai/CLIP#357
No description provided.