# In-Context Black-Box Optimization with Unreliable Feedback

DevFeed: [In-Context Black-Box Optimization with Unreliable Feedback](<https://devfeed.tech/articles/in-context-black-box-optimization-with-unreliable-feedback-118211.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2605.06187>)

Author: Nicolas Samuel Blumer, Julien Martinelli, Samuel Kaski

Published: 2026-11-05T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Hyperparameter optimization](<https://devfeed.tech/topics/hyperparameter-optimization.md>), [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>), [decision-making](<https://devfeed.tech/topics/decision-making.md>)

Tags: [bayesian](<https://devfeed.tech/tags/bayesian.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [feedback](<https://devfeed.tech/tags/feedback.md>), [interpretability](<https://devfeed.tech/tags/interpretability.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [optimization](<https://devfeed.tech/tags/optimization.md>), [real-world](<https://devfeed.tech/tags/real-world.md>), [reliability](<https://devfeed.tech/tags/reliability.md>), [robust](<https://devfeed.tech/tags/robust.md>), [transformer](<https://devfeed.tech/tags/transformer.md>)

## AI overview

FICBO uses a pretrained optimizer to choose black-box optimization queries from both observed results and cheap auxiliary feedback, such as expert suggestions, simulators, predictors, or heuristics. A feedback-aware transformer is pretrained with structured priors spanning informative to misleading sources. At test time, it estimates each source's reliability by comparing its signals with observed objective values, then uses that estimate to guide query selection. The abstract reports effective use of informative feedback and robustness to weak or misleading feedback on synthetic and real-world tasks. It also says empirical investigations examine how the model interprets test-time sources; the supplied text gives no specific task results or performance comparisons.

## Source excerpt

arXiv:2605.06187v2 Announce Type: replace-cross Abstract: Black-box optimization in science and engineering often comes with side information: experts, simulators, pretrained predictors, or heuristics can suggest promising candidates. Such feedback can accelerate search, but can also be biased, input-dependent, or misleading. Feedback-aware Bayesian optimization methods typically handle tasks independently, limiting adaptation across feedback sources. In-context optimizers enable cross-task adaptation, but usually rely only on optimization history 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 structured priors over how auxiliary feedback relates to the objective, and use them to pretrain a feedback-aware transformer across regimes ranging from highly informative to misleading. At test time, the model estimates source reliability in context by comparing observed objective values with auxiliary signals, improving query selection. On synthetic and real-world tasks, FICBO effectively exploits informative feedback while remaining robust to weak or misleading sources. Empirical investigations further illustrate how the model perceives test-time sources, offering insights into its interpretability and decision-making process.