# Hugging Face — How good are LLMs at fixing their mistakes? A chatbot arena experiment with Keras and TPUs

- Company: Hugging Face (huggingface.co)
- Announced: 2024-12-05T00:00:00+00:00
- Subject: Platform
- Source: https://huggingface.co/blog/keras-chatbot-arena
- Record: https://forck.live/items/1787-how-good-are-llms-at-fixing-their-mistakes-a-chatbot-arena-experiment-with

This blog post describes an experiment testing how well LLMs can fix their mistakes when given feedback in plain English, using a simple calendar API scenario. The author built a chatbot arena using Keras, JAX, and TPUs to interact with multiple LLMs simultaneously.

## Evidence

Verbatim from https://huggingface.co/blog/keras-chatbot-arena:

> I decided to run a little test with today's LLMs. A super-simplified one, to see how effectively LLMs fix their mistakes when you point them out to them.

---

Record: https://forck.live/items/1787-how-good-are-llms-at-fixing-their-mistakes-a-chatbot-arena-experiment-with
Catalogue: https://forck.live/llms.txt
Feed: https://forck.live/feed.md