---
title: "Qwen-Flash Gets a Trust Score That Shuts It Down Mid-Optimization"
subtitle: "A new gating mechanism stops LLMs from misleading Bayesian optimization when their confidence is wrong — tested on real molecule benchmarks"
url: "https://alphasignal.ai/news/qwen-flash-gets-a-trust-score-that-shuts-it-down-mid-optimization"
author: "AlphaSignal Newsroom"
author_url: "https://alphasignal.ai/editorial-team"
company: "AlphaSignal"
company_url: "https://alphasignal.ai/company/alphasignal"
published: "2026-06-01T00:00:00.000Z"
updated: "2026-06-09T09:06:45.071Z"
type: "Paper"
topics:
  - "Llms"
  - "Benchmarks"
subtopics:
  - "Hallucinations"
  - "Structured Output"
upvotes: 850
source_url: "https://arxiv.org/abs/2606.01730"
comments: 0
truncated: true
---

# Qwen-Flash Gets a Trust Score That Shuts It Down Mid-Optimization

> A new gating mechanism stops LLMs from misleading Bayesian optimization when their confidence is wrong — tested on real molecule benchmarks

Using an LLM as a smart advisor during optimization sounds like a good idea , until it confidently steers you in the wrong direction. A new paper from Jiangyu Chen and colleagues tackles exactly this failure mode in the context of [multi-objective Bayesian optimization (MOBO)](https://arxiv.org/abs/2606.01730), proposing a dynamic gating system that decides, in real time, how much to trust an LLM's suggestions for each individual optimization objective.

The work sits at the intersection of two fast-moving areas: LLM-guided scientific search and Bayesian optimization for molecule discovery. The core insight is deceptively simple , an LLM might be a great advisor for one property of a molecule (say, drug-likeness) while being actively harmful for another (say, solubility). Treating it as uniformly trustworthy is a mistake.

## The problem with blind trust

Bayesian optimization (BO) is the workhorse for expensive black-box optimization: you have a function you can only evaluate a limited number of times, so you build a probabilistic surrogate model and use it to pick the most promising candidates. Multi-objective Bayesian optimization (MOBO) extends this to handle multiple, often conflicting, expensive-to-evaluate objective functions simultaneously. The goal is to find the **Pareto front** , the set of solutions where you can't improve one objective without hurting another.

Lead optimization in drug discovery involves complex, multi-objective trade-offs, such as simultaneously improving solubility, permeability, and drug-likeness, under strict medicinal chemistry constraints. LLMs have emerged as a natural fit here: they encode vast chemical knowledge and can propose candidate molecules or predict which regions of chemical space are worth exploring. But there's a catch.

Modern LLMs , particularly those aligned by RLHF , can exhibit miscalibrated confidence, complicating uncertainty estimation without substantial validation data. In plain terms: when an LLM says it's 90% confident, that number often doesn't mean what you think it means. Verbalized confidence is systematically overconfident, with nominal 99% confidence intervals covering the true answer only 65% of the time on average. Plugging a miscalibrated prior directly into a BO loop can actively degrade optimization performance.

_Preview only. Full text requires an AlphaSignal Pro account._

## Sources

- [Evidence-Gated LLM Priors for Multi-Objective Bayesian Optimization](https://arxiv.org/abs/2606.01730)
- [AutoLead: An LLM-Guided Bayesian Optimization Framework for Multi-Objective Lead Optimization](https://www.biorxiv.org/content/10.1101/2025.08.19.671029v2.full)
- [Bayesian Optimization of Catalysis with In-Context Learning](https://pubs.acs.org/doi/10.1021/acscentsci.5c02418)
- [Multi-Objective Bayesian Optimization with Independent Tanimoto Kernel Gaussian Processes](https://arxiv.org/pdf/2508.14072)
- [MoleculeNet: a benchmark for molecular machine learning](https://pmc.ncbi.nlm.nih.gov/articles/PMC5868307/)
