GenAI Arena - publication pushed back #1988
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
pcuenca
reviewed
Apr 17, 2024
|
|
||
| In the current ecosystem, evaluating image generation models typically involves scouring through research papers to find relevant metrics, running experiments locally, and comparing results manually. However, much like in text (with the [Chatbot Arena](https://huggingface.co/spaces/lmsys/chatbot-arena-leaderboard)) and speech (with the [TextToSpeech Arena](https://huggingface.co/spaces/TTS-AGI/TTS-Arena)), using an arena system allows both to streamline this process and make sure than models rankings are aligned with human preferences, instead of just relying on automated metrics only offering a partial view of model capabilities. | ||
|
|
||
| We therefore provide a dynamic side-by-side comparison arena, empowering the community to effortlessly generate images, make comparisons, and vote for their preferred model. Our platform is built upon [ImagenHub](https://github.com/TIGER-AI-Lab/ImagenHub), a comprehensive library designed to support inference with various generation and editing models. |
There was a problem hiding this comment.
Suggested change
| We therefore provide a dynamic side-by-side comparison arena, empowering the community to effortlessly generate images, make comparisons, and vote for their preferred model. Our platform is built upon [ImagenHub](https://github.com/TIGER-AI-Lab/ImagenHub), a comprehensive library designed to support inference with various generation and editing models. | |
| Our arena provides a side-by-side comparison of images generated by two models selected randomly from a pool. The community can effortlessly generate images, make comparisons, and vote for their preferred model. Our platform is built upon [ImagenHub](https://github.com/TIGER-AI-Lab/ImagenHub), a comprehensive library designed to support inference with various generation and editing models. |
(Note: voting did not work for me in the current Space. Clicking the small square button in the top-right area of the image gave an error "Share failed". Is there another way to do it?)
There was a problem hiding this comment.
It's possible the authors are adding new models atm, I'll check
|
|
||
| Among the range of editing models available, attention-guided models such as Prompt2prompt and Plug-and-Play (PNP) stand out for their superior performance. These models harness advanced attention control mechanisms to finely edit images in response to user prompts, achieving high consistency with user expectation. | ||
|
|
||
| Instruction-following forward-pass models also excel, offering an impressive balance between image quality and operational efficiency. These models can handle edition without image inversion due to they train with paired annotated data. |
There was a problem hiding this comment.
Suggested change
| Instruction-following forward-pass models also excel, offering an impressive balance between image quality and operational efficiency. These models can handle edition without image inversion due to they train with paired annotated data. | |
| Instruction-following forward-pass models also excel, offering an impressive balance between image quality and operational efficiency. These models can handle edition without image inversion due to being trained with paired annotated data. |
Perhaps links to the models and/or techniques could be helpful for people that want to learn more. Image inversion, for instance, is never defined. No need to explain or cite papers, just a few links could be enough.
clefourrier
commented
Apr 17, 2024
Co-authored-by: Pedro Cuenca <[email protected]>
clefourrier
changed the title
|
Context: authors want to add more results to the blog |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Ready for a light review