# Calibrating Attention-Based Token Pruning for Efficient Multi-Image Synthesis in UMMs

**Authors:** Zimeng Wu, Qijian Tian, Penghui Du, Changqian Yu, Jiaxin Chen, Yunhong Wang  
**Lab:** Klaipėda University · **Published:** 29 Oct 2026  
**Links:** [arXiv](https://arxiv.org/abs/2610.39467) · [PDF](https://arxiv.org/pdf/2610.39467) · [This page](https://omen-bit.github.io/ai-paper-daily/paper/2610.39467.html)

## Key numbers

- **2.09×**: achieving up to 2.09× attention-core speedup

## Abstract

Unified Multimodal Models (UMMs) enable powerful multi-image synthesis but incur substantial computation from long visual contexts. While token pruning offers a direct route to accelerating inference, existing methods are primarily adapted from understanding tasks, suffering from insufficient reference image exploration, confused conditional guidance, and pooling-induced ambiguity that degrades generation quality. To address these issues, we propose CATP, the first training-free context-compression method tailored to UMM generation, which \textbf{C}alibrates the discrepancies of general \textbf{A}ttention-based \textbf{T}oken \textbf{P}runing for better quality-efficiency trade-off. Concretely, building on independent layer-wise pruning during denoising, we introduce Guidance-Consistent Pruning (GCP) to synchronize pruning decisions across conditioning branches, and a Taylor-Calibrated Attention Metric (TCAM) that derives a theoretically grounded second-order correction for pooling-induced bias. In addition, a Fast Scoring Operator (FSO) fuses the calibration into tiled online Softmax to reduce pruning cost. Extensive experiments across representative UMMs and benchmarks demonstrate that CATP consistently outperforms state-of-the-art token-reduction methods, achieving up to 2.09× attention-core speedup.

---

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