MIRG-RL: Multi-Image Reasoning and Grounding with Reinforcement Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zheng, Lihao, Chen, Jiawei, Shen, Xintian, Ma, Hao, Wei, Tao
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911177790980096
author Zheng, Lihao
Chen, Jiawei
Shen, Xintian
Ma, Hao
Wei, Tao
author_facet Zheng, Lihao
Chen, Jiawei
Shen, Xintian
Ma, Hao
Wei, Tao
contents Multi-image reasoning and grounding require understanding complex cross-image relationships at both object levels and image levels. Current Large Visual Language Models (LVLMs) face two critical challenges: the lack of cross-image reasoning capabilities and insufficient cross-image reference reward modeling. To address these issues, we propose a unified framework - Multi-Image Reasoning and Grounding with Reinforcement Learning (MIRG-RL). Specifically, our two-stage training paradigm combines supervised fine-tuning with annotated trajectories and image-aware reinforcement learning optimization, progressively developing multi-image reasoning capabilities. Furthermore, we innovatively propose a method for constructing the trajectory data, which integrates object-level and image-level annotation information, and use this method to generate a lightweight reasoning-enhanced dataset. To effectively resolve cross-image ambiguities, we design an image-aware RL policy with dual reward functions for objects and images. Experiments demonstrate that MIRG-RL achieves state-of-the-art (SOTA) performance in multi-image grounding benchmarks, attaining 64.82% on cross-image reasoning tasks - exceeding the previous best method by 1%. The code and dataset have been released at https://github.com/ZEUS2035/MIRG-RL.
format Preprint
id arxiv_https___arxiv_org_abs_2509_21788
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MIRG-RL: Multi-Image Reasoning and Grounding with Reinforcement Learning
Zheng, Lihao
Chen, Jiawei
Shen, Xintian
Ma, Hao
Wei, Tao
Computer Vision and Pattern Recognition
Multi-image reasoning and grounding require understanding complex cross-image relationships at both object levels and image levels. Current Large Visual Language Models (LVLMs) face two critical challenges: the lack of cross-image reasoning capabilities and insufficient cross-image reference reward modeling. To address these issues, we propose a unified framework - Multi-Image Reasoning and Grounding with Reinforcement Learning (MIRG-RL). Specifically, our two-stage training paradigm combines supervised fine-tuning with annotated trajectories and image-aware reinforcement learning optimization, progressively developing multi-image reasoning capabilities. Furthermore, we innovatively propose a method for constructing the trajectory data, which integrates object-level and image-level annotation information, and use this method to generate a lightweight reasoning-enhanced dataset. To effectively resolve cross-image ambiguities, we design an image-aware RL policy with dual reward functions for objects and images. Experiments demonstrate that MIRG-RL achieves state-of-the-art (SOTA) performance in multi-image grounding benchmarks, attaining 64.82% on cross-image reasoning tasks - exceeding the previous best method by 1%. The code and dataset have been released at https://github.com/ZEUS2035/MIRG-RL.
title MIRG-RL: Multi-Image Reasoning and Grounding with Reinforcement Learning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.21788