Skip to content

BOReFT: Manifold Steering of Language Models for Black-box Optimization

Sep 2026 · arXiv.org · Vol abs/2609.33722 · 0 citations · 50 references
Computer Science

TL;DR

BOReFT is introduced, which learns a compact, low-dimensional space of hidden-state interventions in a frozen language model, and uses this space as the search domain for Bayesian optimization with an external scoring function, and provides a principled new bridge between discrete proposal spaces of LLM-based search and continuous black-box optimization.

Abstract

Language models are increasingly used as proposal models for black-box search, from program optimization to molecular design. Existing approaches typically improve proposals through iterative prompting or parameter updates, offering limited control over how completely and efficiently the model's search space is explored. Continuous optimization methods, such as Bayesian optimization, provide a principled way to search but require a suitable domain to operate over. To address this, we introduce BOReFT, which learns a compact, low-dimensional space of hidden-state interventions in a frozen language model, and uses this space as the search domain for Bayesian optimization with an external scoring function. Empirically, we find that the learned domain spans semantic regions and exhibits smoothness properties that support search. Theoretically, we show that semantic coverage and interpolation control the best score available in the learned space, and that decoding from this space yields a standard stochastic-bandit observation model for adaptive search. We evaluate BOReFT on the interpretable word search task"Semantle"and on three more real-world discovery tasks in de novo molecule property optimization. Compared to strong LLM baselines, BOReFT finds in Semantle a higher number of hidden targets and, on two out of three molecular objectives, achieves higher property scores. Consequently, our method provides a principled new bridge between discrete proposal spaces of LLM-based search and continuous black-box optimization.

View source

Similar papers

#machine learning Preprint Sep 2026

Search at the Cost of Sampling: Nearly-Instant Latent Space Bayesian Optimization

Generative models are increasingly central to many de novo discovery pipelines, in which designs are generated at scale and filtered through virtual screens to determine a set of candidates to experimentally validate. While Bayesian optimization (BO) is a natural fit for this setting, as it uses past evaluations to gui...

Donney Fan, Colin Doumont, Aleksandra Kalisz et al. · 4 citations · ⚡1

Bayesian Optimization with Rich Auxiliary Information via LLMs

Two methods are introduced, AuxBO-Evolve, which uses auxiliary information to evolve beliefs over the location of the optimum, and AuxBO-Vicinity, which uses it to locally guide the acquisition function, which consistently improve optimization performance and outperform standard BO and existing LLM-based approaches.

Tejus Gupta, Efe Mert Karagözlü, Rohit Sonker et al. · 0 citations

Behavioral Foundation Models for Quality Diversity

The effectiveness of the BFM-QD framework is shown, benefiting from the synergy between dimensionality reduction of the search space and offline pretraining from diverse behavioral data, which positions BFMs as a general-purpose backbone for QD optimization.

Nazim Bendib, Nicolas Perrin-Gilbert, Olivier Sigaud · 0 citations
#natural language process... Preprint Aug 2026

Closed-Loop Bayesian Molecular Inverse Design with Semantic LLM Surrogates

Experiments show that BoMolLLM improves over one-shot prompting, is competitive with or stronger than GP-based BO baselines, and reveals a domain-dependent interface: reference-only transfer works best for binary drug targets, while adding a concise surrogate summary is more beneficial for continuous material.

Yao-Yao Xu, Xinjian Zhao, Xiao-Zhuang Song et al. · 0 citations
#machine learning Preprint Sep 2026

Agentic Search Spaces for Tabular Machine Learning

This study suggests that LLM agents can provide practical value for tabular ML by expanding the design space, and suggests that the two strongest agentic ensembles surpass the best AutoGluon ensemble of conventional models.

Renat Sergazinov, Artem Chistyakov, Sergey Pankevich et al. · 0 citations
Preprint Oct 2026

Bayesian Optimization in Sequence-to-Architecture Latent Space for Zero-Shot NAS

Zero-shot Neural Architecture Search removes the prohibitive cost of traditional NAS, but its search process is typically based on the evolutionary algorithm (EA); lacking an explicit model of the objective, it often resorts to a near-random search through mutation. Bayesian Optimization offers a principled alternative...

Ondrej Týbl, Lukás Neumann · 0 citations

Related blog posts

MIT News · Artificial Intelligence Nov 17, 2026

Learning the language of machine learning

By studying the computational foundations of language learning, Associate Professor Jacob Andreas builds intelligent systems that can communicate effectively with humans.

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.