AI Paper Atlas

Research map / Probabilistic and scientific modeling / Bayesian optimization

Thompson Sampling via Fine-Tuning of LLMs

Nicolas Menet, Aleksandar Terzic, Michael Hersche, Andreas Krause, Abbas Rahimi

ICLR 2026 · Probabilistic and scientific modeling › Bayesian optimization · approach: Local and high-dimensional Bayesian optimization · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add model fine-tuning to generationToSFiT · Protein Search wall-clocklatency per step: 19% overhead (1.08 s generation, 1.29 s total)Hurt
add optimism-driven fine-tuned sampling policy vs unguided LLM generationToSFiT · FAQ Response Refinement, Protein Search, Quantum Circuit Designbest-seen rewardImproved
add optimism instead of undirected exploration (actor critics)ToSFiT · three tasksexploration stabilityImproved
use parameter updates instead of in-context learning (FIBO)ToSFiT · three tasksrewardImproved
increase model size (ablation)ToSFiT · FAQ Response Refinement vs Protein Search and Quantum CircuitrewardMixed
increase batch size up to 16 (ablation)batched ToSFiT · Protein Searchsample vs iteration efficiency: up to 16 proteins in parallelMixed
add more gradient ascent steps per round (ablation)ToSFiT · computational efficiency studysample efficiencyImproved
mask number of qubits in the prompt (weak context) (ablation)ToSFiT · Quantum Circuit DesignrewardHurt
use a large learning rate (ablation)ToSFiT · fine-tuning learning rate studyrewardHurt
add entropy-like term sqrt(-2 ln pi) sigma to objectiveToSFiT · generation diversitydiversityImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy