Saved in:
Bibliographic Details
Main Authors: Xue, Yu, Qu, Haoxuan, Li, Zhuoling, Lou, Yihang, Bai, Yan, Rahmani, Hossein, Liu, Jun
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2606.02518
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911742323326976
author Xue, Yu
Qu, Haoxuan
Li, Zhuoling
Lou, Yihang
Bai, Yan
Rahmani, Hossein
Liu, Jun
author_facet Xue, Yu
Qu, Haoxuan
Li, Zhuoling
Lou, Yihang
Bai, Yan
Rahmani, Hossein
Liu, Jun
contents Fine-grained image classification (FGIC) has broad applications and has attracted significant research attention. In this paper, we explore a novel paradigm for solving FGIC by proposing \textbf{ToolFG}, the first tool-integrated MLLM-based framework tailored to FGIC. ToolFG enables MLLMs to autonomously and flexibly use external tools during the reasoning process, actively interact with images, and collect verifiable visual cues for distinguishing highly similar categories in a more \textit{reliable} and \textit{well-grounded} manner. To equip the model with such tool-use ability, we design a novel \textbf{MCTS-guided tool-use knowledge distillation mechanism}, which effectively mines tool-use- and FGIC-relevant knowledge from advanced proprietary MLLMs for model training. Furthermore, we propose a \textbf{model-tool co-evolution mechanism} that jointly refines the toolset and the model's tool-use policy, driving them toward a mutually adapted and FGIC-specialized state. Extensive experiments demonstrate the effectiveness of our framework.
format Preprint
id arxiv_https___arxiv_org_abs_2606_02518
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle ToolFG: Towards Well-Grounded Fine-Grained Image Classification
Xue, Yu
Qu, Haoxuan
Li, Zhuoling
Lou, Yihang
Bai, Yan
Rahmani, Hossein
Liu, Jun
Computer Vision and Pattern Recognition
Fine-grained image classification (FGIC) has broad applications and has attracted significant research attention. In this paper, we explore a novel paradigm for solving FGIC by proposing \textbf{ToolFG}, the first tool-integrated MLLM-based framework tailored to FGIC. ToolFG enables MLLMs to autonomously and flexibly use external tools during the reasoning process, actively interact with images, and collect verifiable visual cues for distinguishing highly similar categories in a more \textit{reliable} and \textit{well-grounded} manner. To equip the model with such tool-use ability, we design a novel \textbf{MCTS-guided tool-use knowledge distillation mechanism}, which effectively mines tool-use- and FGIC-relevant knowledge from advanced proprietary MLLMs for model training. Furthermore, we propose a \textbf{model-tool co-evolution mechanism} that jointly refines the toolset and the model's tool-use policy, driving them toward a mutually adapted and FGIC-specialized state. Extensive experiments demonstrate the effectiveness of our framework.
title ToolFG: Towards Well-Grounded Fine-Grained Image Classification
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2606.02518