---
format: "aidr-story-markdown/v1"
id: "732714e7e6ffa514ace7b986b01d1705862c239aa19880447940c8de7071fd59"
canonical_url: "https://aidr.today/732714e7?lang=en"
title: "Diffusion Tree Sampling for Inference Time Adaptation of Crystal Generative Models"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-11-16T05:00:00.000Z"
category: "Research"
topics: ["inference"]
source_urls: ["https://arxiv.org/abs/2611.17462"]
summary: "Diffusion Tree Sampling (DTS) adapts crystal generative models at inference time using a Monte Carlo tree search over the denoising trajectory. DTS can use existing property-conditional generative models as guided proposals inside DTS without distorting the target distribution. DTS was extended to multi-objective sampling to produce samples spanning an entire Pareto front."
---

# Diffusion Tree Sampling for Inference Time Adaptation of Crystal Generative Models

> [Open the canonical story](<https://aidr.today/732714e7?lang=en>)

**Published:** 2026-11-16T05:00:00.000Z
**Category:** Research
**Topics:** inference

## Summary

Diffusion Tree Sampling \(DTS\) adapts crystal generative models at inference time using a Monte Carlo tree search over the denoising trajectory\. DTS can use existing property\-conditional generative models as guided proposals inside DTS without distorting the target distribution\. DTS was extended to multi\-objective sampling to produce samples spanning an entire Pareto front\.

## Sources

- [Story source](<https://arxiv.org/abs/2611.17462>)

