Skip to content

Art request - LMArena Case Study #14322

Description

@cleo-pleurodon

Outline

New customer case study in the works with LMArena !
LMArena evaluates LLMs through side-by-side, blind comparisons. Users see two anonymous responses to the same prompt and vote on which one is stronger.

Concept

Important: LMArena is doing a rebrand at the end of January and will send us a new logo. I'm not sure what else will change about their branding but will keep you posted as I learn more)

For the image, something with an AI battle / single combat / gladiator vibe would be fun, but I'm cool with wherever you want to take this.
Here's a few key components of their product if it's helpful:

  • There's a leaderboard for each modality (text, video, code...)
  • The comparison battles are always 1 on 1
  • The AI outputs are always anonymous until you vote (to prevent biased voting)
  • Human preference data is actually really useful to frontier labs (OpenAI, Meta, Anthropic...) to benchmark models and test unreleased variants
Image Image

Deadline & Priority

We're planning to publish the first week of February, so end of January would be much appreciated!

Technical requirements

Standard case study image size

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Labels

Type

No type

Projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions