---
format: "aidr-story-markdown/v1"
id: "cc3d0ac48231448ce58666765641670f54b8dd433c8df85dffcc57804ac94ee3"
canonical_url: "https://aidr.today/cc3d0ac4?lang=en"
title: "Limits of Confidence in Diffusion"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-02T00:00:00.000Z"
category: "Research"
topics: ["inference"]
source_urls: ["https://machinelearning.apple.com/research/limits-confidence-diffusion"]
summary: "Authors Russ Webb, Amitis Shidani, Alice Bizeul, and Dan Busbridge argue that discrete diffusion methods generate sequences by writing multiple token positions per step from per-position distributions, and show that a step matches the training distribution only when written positions are conditionally independent given already fixed tokens."
---

# Limits of Confidence in Diffusion

> [Open the canonical story](<https://aidr.today/cc3d0ac4?lang=en>)

**Published:** 2026-10-02T00:00:00.000Z
**Category:** Research
**Topics:** inference

## Summary

Authors Russ Webb, Amitis Shidani, Alice Bizeul, and Dan Busbridge argue that discrete diffusion methods generate sequences by writing multiple token positions per step from per\-position distributions, and show that a step matches the training distribution only when written positions are conditionally independent given already fixed tokens\.

## Sources

- [Story source](<https://machinelearning.apple.com/research/limits-confidence-diffusion>)

