# Probing Local Conditional Fit for Diffusion Model Membership Inference

**Authors:** ByungHee Cha, Naoki Murata, Yuhta Takida, Jong Chul Ye, Yuki Mitsufuji  
**Lab:** KAIST · **Published:** 2 Nov 2026  
**Links:** [arXiv](https://arxiv.org/abs/2611.02578) · [PDF](https://arxiv.org/pdf/2611.02578) · [This page](https://omen-bit.github.io/ai-paper-daily/paper/2611.02578.html)

## Abstract

Membership inference determines whether a candidate example was used to train a text-to-image diffusion model and is an important tool for auditing training-data use. We identify a directional membership signal in \emph{conditional prediction shifts}, the changes in predicted noise under alternative text conditions at a fixed noisy input: whether movement toward the alternative prediction initially lowers or raises the denoising loss from its value under the original caption. Members exhibit less room for such local improvement. To quantify this response, we propose \emph{LoFIT} (Local Conditional Fit), which measures the initial response of the denoising loss along normalized prediction shifts and aggregates it across multiple text conditions and diffusion states. LoFIT requires no target-model gradients and can be computed using only forward queries to the denoiser. Evaluations on four datasets show improved membership discrimination across fine-tuned and pretrained models. LoFIT distinguishes members early in fine-tuning, while the naive denoising-loss baseline remains near chance. Controlled comparisons show that the membership signal is concentrated in the first-order response, with little discrimination from prediction-shift magnitude alone.

---

Notes written by AI from the paper, numbers copied from the abstract. Check the paper before you rely on them. From AI Papers Daily.