arXiv cs.AI· Dohyun Kim, Junyong Lhim, Kwanyong Park·· 11 小时前AI 评分26
MPRBench：评估视觉语言模型多模态个性化推理的系统性基准
A Systematic Benchmark for Evaluating Multimodal Personalized Reasoning of Vision-Language Models
AI 导读
研究者提出 MPRBench，一个专门评估多模态大模型（MLLMs）个性化推理能力的基准，包含 12 个子任务，并支持对代表性错误案例的可解释分析。实验发现，当前个性化 MLLMs 仍难以完成个性化推理，主要挑战包括对个性化概念识别不准确以及对无关个人信息敏感。该工作已被 WACV 2027 接收，代码与数据已公开。
正文 · 原文
Abstract:People often make decisions by holistically reasoning over heterogeneous forms of personal information. In this paper, we study the relatively underexplored capability of multimodal personalized reasoning in multimodal large language models (MLLMs). To this end, we introduce \texttt{MPRBench}, a novel and comprehensive benchmark specifically designed to evaluate personalized reasoning across a wide range of tasks and real-world challenges. \texttt{MPRBench} consists of 12 sub-tasks and additionally supports interpretable analysis of representative error cases. Through extensive experiments, we find that current personalized MLLMs still struggle with personalized reasoning due to several key challenges, including inaccurate recognition of personalized concepts and sensitivity to irrelevant personal information. We hope that MPRBench will stimulate further research on personalized reasoning in MLLMs. Code and data are available at this https URL
| Comments: | Accepted to WACV 2027 |
| Subjects: | Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2611.06259 [cs.AI] |
| (or arXiv:2611.06259v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.06259 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Dohyun Kim [view email]
[v1]
Wed, 4 Nov 2026 10:08:26 UTC (9,286 KB)
来源：arXiv cs.AI · arxiv.org