[2605.06187] In-Context Black-Box Optimization with Unreliable Feedback

Computer Science > Machine Learning arXiv:2605.06187 (cs) [Submitted on 7 May 2026] Title:In-Context Black-Box Optimization with Unreliable Feedback Authors:Nicolas Samuel Blumer, Julien Martinelli, Samuel Kaski View a PDF of the paper titled In-Context Black-Box Optimization with Unreliable Feedback, by Nicolas Samuel Blumer and 2 other authors View PDF HTML (experimental) Abstract:Black-box optimization in science and engineering often comes with side information: experts, simulators, pretrained predictors, or heuristics can suggest which candidates look promising. This information can accelerate search, but it can also be biased, input-dependent, or misleading. Feedback-aware BO methods typically handle one task at a time, limiting their ability to generalize over multiple sources of feedback. In-context optimizers address cross-task adaptation, but usually assume that optimization history is the only available signal at test time. We study feedback-informed in-context black-box optimization (FICBO), where a pretrained optimizer conditions on both the observed history and cheap auxiliary feedback for the current candidate set. We introduce a structured feedback prior that models how feedback sources vary in their access, relevance, and distortion relative to the true objective, and use it to pretrain a feedback-aware transformer. At test time, the model estimates source reliability in context by comparing observed objective values with auxiliary signals, improving query...