SOURCE-LINKED INTELLIGENCE

Online Learning with LLM Experts from Limited Feedback

arXiv · AI, language, vision and robotics · article · Sep 5, 2026 · UTC

We study adaptive routing of prompts to large language model (LLM) experts to maximize response quality in an online setting with limited feedback. We formulate it as a bandit problem with $K$ actions that represent experts and $d$ features that encode prompts, over a horizon of $T$ rounds. We propose algorithms that strategically select and observe rewards to minimize regret. In the full-information setting, we achieve a regret of $\tilde{O}(d T / \sqrt{m})$, while in the bandit setting we achieve $\tilde{O}(d T \sqrt{K / m})$, where $m \ll T$ is a budget on feedback. Our experiments show tha

Read original source ↗ Open in workspace

recordType

paper

region

Global