ReGraP-LLaVA: Reasoning enabled Graph-based Personalized Large Language and Vision Assistant

TL;DR

ReGraP-LLaVA introduces a novel dataset and model for personalized multi-modal reasoning, enabling structured relational understanding among personalized concepts, and achieves state-of-the-art performance on diverse reasoning tasks.

Contribution

The paper presents ReGraP, a new dataset with structured reasoning pathways, and ReGraP-LLaVA, a model trained on this data to enhance personalized relational reasoning in multimodal tasks.

Findings

ReGraP-LLaVA outperforms existing models on the ReGraP benchmark.

The dataset enables training models to reason over relations among personalized concepts.

Graph prompting methods improve the alignment of knowledge graphs within the model.

Abstract

Recent advances in personalized MLLMs enable effective capture of user-specific concepts, supporting both recognition of personalized concepts and contextual captioning. However, humans typically explore and reason over relations among objects and individuals, transcending surface-level information to achieve more personalized and contextual understanding. To this end, existing methods may face three main limitations: Their training data lacks multi-object sets in which relations among objects are learnable. Building on the limited training data, their models overlook the relations between different personalized concepts and fail to reason over them. Their experiments mainly focus on a single personalized concept, where evaluations are limited to recognition and captioning tasks. To address the limitations, we present a new dataset named ReGraP, consisting of 120 sets of personalized knowledge. Each set includes images, KGs, and CoT QA pairs derived from the KGs, enabling more structured and sophisticated reasoning pathways. We propose ReGraP-LLaVA, an MLLM trained with the corresponding KGs and CoT QA pairs, where soft and hard graph prompting methods are designed to align KGs within the model's semantic space. We establish the ReGraP Benchmark, which contains diverse task types: multiple-choice, fill-in-the-blank, True/False, and descriptive questions in both open- and closed-ended settings. The proposed benchmark is designed to evaluate the relational reasoning and knowledge-connection capability of personalized MLLMs. We conduct experiments on the proposed ReGraP-LLaVA and other competitive MLLMs. Results show that the proposed model not only learns personalized knowledge but also performs relational reasoning in responses, achieving the SoTA performance compared with the competitive methods. All the codes and datasets are released at: https://github.com/xyfyyds/ReGraP.