← Back to all news
arXiv Machine Learning

LLINBO: Trustworthy LLM-in-the-Loop Bayesian Optimization

LLINBO is a framework that integrates Large Language Models (LLMs) into Bayesian Optimization (BO) by deciding when to use LLM suggestions and when to fall back on classical BO. It introduces three interaction mechanisms, each with a no‑harm theoretical guarantee of sublinear cumulative regret for common kernels, without requiring performance assumptions on the LLM. Experiments show that LLINBO improves the reliability of LLM‑based BO and outperforms classical BO in both simulations and a real‑world 3D printing task.

Related stories

arXiv AI
6d ago

Rethinking Bayesian Optimization for Co-Optimizing LLM Training Configurations

The paper introduces JoBS, a Bayesian optimization framework that co‑optimizes data and model configurations for large language model training without requiring full training runs at each iteration. JoBS first allocates part of the budget to train a scaling‑law‑inspired performance predictor using only a few training steps, then uses this predictor to guide subsequent BO iterations, thereby exploring many more configurations within the same budget. The authors analyze JoBS’s regret, derive optimal budget splits, and demonstrate that it outperforms both independent optimization and existing multi‑fidelity BO baselines across diverse downstream tasks.

By Zhiliang Chen, Alfred Wei Lun Leong, Shao Yong Ong, Apivich Hemachandra, Gregory Kang Ruey Lau, Chuan-Sheng Foo, Zhengyuan Liu, Nancy F. Chen, Bryan Kian Hsiang Low
llmsfine-tuning
More like this →
arXiv AI
Oct 28

BoLT: A Benchmark to Democratize Black-box Optimization Research for Expensive LLM Tasks

BoLT is a new benchmark designed to make black-box optimization research for large language model (LLM) tasks more accessible. It presents a suite of realistic LLM optimization problems—covering multi‑fidelity, multi‑objective, heteroscedastic noise, and high‑dimensional search spaces—grounded in thousands of real experiments and represented by lightweight surrogate models. The authors evaluate a wide range of Bayesian and other black‑box optimization methods on BoLT, finding that some BO techniques consistently outperform others and revealing gaps in current methods for LLM optimization.

By Ruth Wan Theng Chew, Zhiliang Chen, Apivich Hemachandra, Bryan Kian Hsiang Low
llmsbenchmarks
More like this →
arXiv Machine Learning
Sep 4

No-Regret Bayesian Optimization with Finite-Library Input-Warped Kernels

The paper introduces Finite-Library Input-Warped Bayesian Optimization (FLIWBO), a method that selects input warps from a finite library to adapt the geometry used by Gaussian‑process Bayesian optimization. FLIWBO maintains high‑probability convergence guarantees while improving sample efficiency on problems where raw coordinates poorly match the objective’s geometry, such as log‑scaled hyperparameters or localized peaks. Experiments on synthetic benchmarks, Fashion‑MNIST hyperparameter tuning, and a 20‑dimensional multi‑agent system design demonstrate that FLIWBO‑UCB outperforms raw‑coordinate GP‑UCB and other methods with regret guarantees, especially under misspecified geometry.

By Edvin Ketabati Augustinsson, Robert A. Bridges
agentsbenchmarks
More like this →
arXiv Machine Learning
Aug 3

Frugal Bayesian Optimization: Scalable Surrogates for Data- and Resource-Limited Discovery

arXiv:2607. 29225v1 Announce Type: new Abstract: Bayesian Optimization (BO) is widely adopted for data-efficient optimization in scientific and engineering applications, yet its computational cost is rarely evaluated alongside optimization performance.

By Panagiotis Krokidas, Christoforos Rekatsinas, Vassilis Sioros, Grigorios M. Chatziathanasiou, Efi-Maria Papia, George Giannakopoulos
roboticsbenchmarks
More like this →
arXiv Machine Learning
Oct 23

SABLE: Scalable Affine Bayesian Latent-space Exploration

SABLE is a Bayesian optimisation method designed for high‑dimensional latent spaces where the latent dimension far exceeds the evaluation budget. It uses a Bayesian affine surrogate that decomposes each query into an in‑span component optimized by expected improvement and an off‑span component governed by a scalar norm, with a finite‑horizon schedule to avoid under‑exploration. A recompression scheme reduces the number of pseudo‑observations to ≠O(M), keeping refitting costs manageable while achieving high sample efficiency and rapid decision times on synthetic and latent‑space optimisation tasks.

By Erik Bodin, Lisa Maria Kreusser, Donney Fan, Henry Moss, Geoff Pleiss, Nicolas Durrande
benchmarks
More like this →