AlpacaEval
Emerging5papers using it
20,423HF downloads
65HF likes
2024first seen
Data for alpaca_eval, which aims to help automatic evaluation of instruction-following models
🤗 Hugging Face⚖ cc-by-nc-4.0
Papers using AlpacaEval (5)
- References Improve LLM Alignment in Non-Verifiable DomainsOnline Rubrics Elicitation from Pairwise ComparisonsPretrain Value, Not Reward: Decoupled Value Policy OptimizationPost-hoc Reward Calibration: A Case Study on Length BiasSentence-level Reward Model can Generalize Better for Aligning LLM from Human Preference