DaoyuanLi2816/Kaggle-4th-Place-Solution-LMSYS-Chatbot-Arena-Human-Preference-Predictions

4th Place Solution for the Kaggle Competition: LMSYS - Chatbot Arena Human Preference Predictions

36
/ 100
Emerging

This solution predicts which chatbot response humans will prefer in head-to-head conversations. It takes transcripts of user prompts and two chatbot responses as input and outputs a prediction of which response (A or B) is better, or if it's a tie. This is designed for AI researchers and developers who need to understand and improve how well their language models align with human preferences.

170 stars. No commits in the last 6 months.

Use this if you need to automatically predict human preferences for chatbot responses to improve your AI's conversational quality.

Not ideal if you are looking for a system to generate chatbot responses directly, rather than evaluate existing ones.

AI-evaluation chatbot-preference LLM-alignment natural-language-processing human-in-the-loop-AI
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 10 / 25

How are scores calculated?

Stars

170

Forks

11

Language

Jupyter Notebook

License

MIT

Last pushed

Oct 24, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/DaoyuanLi2816/Kaggle-4th-Place-Solution-LMSYS-Chatbot-Arena-Human-Preference-Predictions"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.