Conditioning noise is a free regularizer for LoRA fine-tuning: no pathology encoder required for diffusion-based artifact detection in histopathology

2026-09-11

AI Summary

研究通过在LoRA微调过程中引入随机高斯嵌入作为条件，无需额外编码器，有效提高了基于扩散的病理学图像伪影检测中清洁与伪影区域的区分度。

Abstract

Diffusion-based artifact detectors score whole-slide image patches by reconstruction error under a model fine-tuned on clean tissue. We show that conditioning this fine-tuning on random Gaussian embeddings -- resampled at every step from approx. 200 KB of precomputed embedding statistics, with no encoder, no cache, and no change to inference -- consistently widens the clean/artifact separation. A four-step ablation chain shows the benefit requires neither content (shuffled real embeddings), provenance (synthetic Gaussians), a tuned intensity (flat across an 8x variance range), nor per-patch identity (fresh per-step noise); a LoRA-dropout control shows the conditioning pathway specifically, not generic weight perturbation, carries the effect. Patch-level gains of +0.25-0.48 Cohen's d replicate across nine trainings; honest leave-one-slide-out evaluation clears a pre-registered bar in 2/2 seeds; and two pre-registered external endpoints on a 281-case set confirm pooled Delta F1 = +0.0073 (95% CI) and +0.0129 (97.5% CI, two-look corrected). We release the full evaluation protocol, including measured seed noise and selection-optimism pricing.

Problem

• diffusion-based artifact detection
• histopathology
• reconstruction error
• Gaussian embeddings
• LoRA fine-tuning

Innovation

• Conditioning Noise
• LoRA Fine-tuning
• Diffusion-based Artifact Detection
• Random Gaussian Embeddings
• Regularization