Saved in:
Bibliographic Details
Main Authors: Jin, Xin, Zhong, Yichuan, Tian, Yapeng
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2601.08011
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915825392287744
author Jin, Xin
Zhong, Yichuan
Tian, Yapeng
author_facet Jin, Xin
Zhong, Yichuan
Tian, Yapeng
contents Current text-conditioned diffusion editors handle single object replacement well but struggle when a new object and a new style must be introduced simultaneously. We present Twin-Prompt Attention Blend (TP-Blend), a lightweight training-free framework that receives two separate textual prompts, one specifying a blend object and the other defining a target style, and injects both into a single denoising trajectory. TP-Blend is driven by two complementary attention processors. Cross-Attention Object Fusion (CAOF) first averages head-wise attention to locate spatial tokens that respond strongly to either prompt, then solves an entropy-regularised optimal transport problem that reassigns complete multi-head feature vectors to those positions. CAOF updates feature vectors at the full combined dimensionality of all heads (e.g., 640 dimensions in SD-XL), preserving rich cross-head correlations while keeping memory low. Self-Attention Style Fusion (SASF) injects style at every self-attention layer through Detail-Sensitive Instance Normalization. A lightweight one-dimensional Gaussian filter separates low- and high-frequency components; only the high-frequency residual is blended back, imprinting brush-stroke-level texture without disrupting global geometry. SASF further swaps the Key and Value matrices with those derived from the style prompt, enforcing context-aware texture modulation that remains independent of object fusion. Extensive experiments show that TP-Blend produces high-resolution, photo-realistic edits with precise control over both content and appearance, surpassing recent baselines in quantitative fidelity, perceptual quality, and inference speed.
format Preprint
id arxiv_https___arxiv_org_abs_2601_08011
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle TP-Blend: Textual-Prompt Attention Pairing for Precise Object-Style Blending in Diffusion Models
Jin, Xin
Zhong, Yichuan
Tian, Yapeng
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Multimedia
Current text-conditioned diffusion editors handle single object replacement well but struggle when a new object and a new style must be introduced simultaneously. We present Twin-Prompt Attention Blend (TP-Blend), a lightweight training-free framework that receives two separate textual prompts, one specifying a blend object and the other defining a target style, and injects both into a single denoising trajectory. TP-Blend is driven by two complementary attention processors. Cross-Attention Object Fusion (CAOF) first averages head-wise attention to locate spatial tokens that respond strongly to either prompt, then solves an entropy-regularised optimal transport problem that reassigns complete multi-head feature vectors to those positions. CAOF updates feature vectors at the full combined dimensionality of all heads (e.g., 640 dimensions in SD-XL), preserving rich cross-head correlations while keeping memory low. Self-Attention Style Fusion (SASF) injects style at every self-attention layer through Detail-Sensitive Instance Normalization. A lightweight one-dimensional Gaussian filter separates low- and high-frequency components; only the high-frequency residual is blended back, imprinting brush-stroke-level texture without disrupting global geometry. SASF further swaps the Key and Value matrices with those derived from the style prompt, enforcing context-aware texture modulation that remains independent of object fusion. Extensive experiments show that TP-Blend produces high-resolution, photo-realistic edits with precise control over both content and appearance, surpassing recent baselines in quantitative fidelity, perceptual quality, and inference speed.
title TP-Blend: Textual-Prompt Attention Pairing for Precise Object-Style Blending in Diffusion Models
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Multimedia
url https://arxiv.org/abs/2601.08011