Don't Send What You Don't Need: Question-Guided Token Pruning as a Privacy Defense for Vision-Language Models

Paper recorded by Signals 4 on 2026-09-14 in cs.CV. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-14 on arXiv · recorded by Signals 4 on 2026-09-15

Category: cs.CV · 计算机视觉 · first seen 2026-09-15

Abstract

Visual Question Answering (VQA) with Vision-Language Models (VLMs) is increasingly used in privacy-sensitive and bandwidth-constrained settings. Federated Learning (FL), Split Learning (SL), and U-Shaped Split Learning (USL) keep raw data local, but transmitting all visual tokens across a model partition remains costly and can expose private information. We propose QPriv-VL, a question-guided, pri

Read on arXiv →