Reinforcement Learning from Human Feedback (RLHF) has become the standard for aligning Large Language Models (LLMs), yet its efficacy is bottlenecked by the high cost of acquiring preference data, especially in low-resource and expert domains. To address this, we introduce ActiveUltraFeedback, a modular active learning pipeline that leverages uncertainty estimates to dynamically identify the most informative responses for annotation. Our pipeline facilitates the systematic evaluation of standard response selection methods alongside Double Reverse Thompson Sampling (DRTS) and DeltaUCB, two novel methods prioritizing response pairs with large predicted quality gaps, leveraging recent results showing that such pairs provide good signals for fine-tuning. Our experiments demonstrate that ActiveUltraFeedback yields high-quality datasets that lead to significant improvements in downstream performance, notably achieving comparable or superior results with as little as one-sixth of the annotated data relative to static baselines.

@inproceedings{melikidze2026activeultrafeedbackefficientpreferencedata,
      title={ActiveUltraFeedback: Efficient Preference Data Generation using Active Learning}, 
      author={Davit Melikidze* and Marian Schneider* and Jessica Lam* and Martin Wertich* and Ido Hakimi and Barna Pásztor and Andreas Krause},
      year={2026},
	  booktitle = {International Conference on Machine Learning (ICML)},
	  month={July},
	  pdf = {https://arxiv.org/pdf/2603.09692},
	  blog={https://lasgroup.github.io/rlhf/},
}