Preprint · 2026 · Open access
Mags-RL: Wearing Multimodal LLMs a Magnifying Glass via Agentic Reinforcement Learning For Complex Scene Reasoning
arXiv (Cornell University) · Published 27 May 2026 · DOI 10.48550/arxiv.2605.27960
In brief
“Mags-RL: Wearing Multimodal LLMs a Magnifying Glass via Agentic Reinforcement Learning For Complex Scene Reasoning” is a 2026 preprint by Xuanzhao Dong and 13 co-authors, published in arXiv (Cornell University). OpenAlex counts 0 citations as of 19 November 2026. It is open access: the full text is free to read at arXiv (Cornell University).
- Citations
- 0
- References
- 52
- Authors
- 14
Abstract
Despite their popularity and success, Multimodal Large Language Models (MLLMs) often struggle to interpret images accurately, which limits their reasoning capability in complex scenarios (e.g., high object density and complex background clutter). Prior work mainly addresses this limitation by incorporating explicit visual cues like bounding boxes that require extra annotations. In addition, the resulting low-resolution crops often miss fine-grained details that MLLMs require for accurate reasoning. Therefore, we propose Mags-RL, an Agentic Reinforcement Learning (RL) framework that equips MLLMs with an external super-resolution "magnifying glass" agent for high-resolution fine-grained inspection. Specifically, the model performs two-round reasoning: in the first round, it generates an initial rationale and autonomously identifies regions of interest without relying on additional annotations; in the second round, it invokes a super-resolution agent to crop and upscale those regions, then revisits and verifies its earlier reasoning to produce the final answer. We also introduce a novel curriculum learning strategy that enables data-efficient RL training, needing as few as only 40 training samples to achieve reasonable performance. Experiments on VSR, TallyQA, and GQA subsets show its superior performance against recent strong competing methods, demonstrating high-quality reasoning with precise visual grounding. Code and weights will be released soon.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Dong, X., Zhu, W., Qiu, P., Chen, X., Yu, X., Li, X., Wang, Z., Tang, S., Li, G., Xiong, Y., Wang, H., Chen, Y., Tiwari, P., & Wang, Y. (2026). Mags-RL: Wearing Multimodal LLMs a Magnifying Glass via Agentic Reinforcement Learning For Complex Scene Reasoning. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2605.27960
MLA 9
Dong, Xuanzhao, et al. “Mags-RL: Wearing Multimodal LLMs a Magnifying Glass via Agentic Reinforcement Learning For Complex Scene Reasoning.” arXiv (Cornell University), 2026. https://doi.org/10.48550/arxiv.2605.27960.
Chicago
Dong, Xuanzhao, Wenhui Zhu, Peijie Qiu, Xiwen Chen, Xiaobing Yu, Xin Li, Zhipeng Wang, et al. 2026. “Mags-RL: Wearing Multimodal LLMs a Magnifying Glass via Agentic Reinforcement Learning For Complex Scene Reasoning.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2605.27960.
Harvard
Dong, X. et al. (2026) ‘Mags-RL: Wearing Multimodal LLMs a Magnifying Glass via Agentic Reinforcement Learning For Complex Scene Reasoning’, arXiv (Cornell University). doi:10.48550/arxiv.2605.27960.
IEEE
X. Dong et al., “Mags-RL: Wearing Multimodal LLMs a Magnifying Glass via Agentic Reinforcement Learning For Complex Scene Reasoning,” arXiv (Cornell University), 2026, doi: 10.48550/arxiv.2605.27960.
BibTeX
@misc{dong2026mags,
title = {{Mags-RL: Wearing Multimodal LLMs a Magnifying Glass via Agentic Reinforcement Learning For Complex Scene Reasoning}},
author = {Dong, Xuanzhao and Zhu, Wenhui and Qiu, Peijie and Chen, Xiwen and Yu, Xiaobing and Li, Xin and Wang, Zhipeng and Tang, Shao and Li, Gen and Xiong, Yujian and Wang, Hao and Chen, Yanxi and Tiwari, Prayag and Wang, Yalin},
howpublished = {arXiv (Cornell University)},
year = {2026},
doi = {10.48550/arxiv.2605.27960},
url = {https://doi.org/10.48550/arxiv.2605.27960}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Topic and keywords
Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Multimodal Machine Learning Applications
- multimodal large language models
- agentic reinforcement learning
- image magnification
- visual grounding
- region of interest detection
- curriculum learning
- data-efficient reinforcement learning
- visual question answering
Work with this paper
Most-cited papers in Computer Science
- ImageNet classification with deep convolutional neural networksAlex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton · Communications of the ACM · 2017 · 114,726 citations
- Long Short-Term MemorySepp Hochreiter, Jürgen Schmidhuber · Neural Computation · 1997 · 109,097 citations
- A Mathematical Theory of CommunicationClaude E. Shannon · Bell System Technical Journal · 1948 · 91,736 citations
- Deep learningYann LeCun, Yoshua Bengio, Geoffrey E. Hinton · Nature · 2015 · 90,226 citations
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksShaoqing Ren, Kaiming He, Ross Girshick and 1 more · IEEE Transactions on Pattern Analysis and Machine Intelligence · 2016 · 74,143 citations
Questions about this paper
How do I cite this paper in APA style?
Dong, X., Zhu, W., Qiu, P., Chen, X., Yu, X., Li, X., Wang, Z., Tang, S., Li, G., Xiong, Y., Wang, H., Chen, Y., Tiwari, P., & Wang, Y. (2026). Mags-RL: Wearing Multimodal LLMs a Magnifying Glass via Agentic Reinforcement Learning For Complex Scene Reasoning. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2605.27960
How many times has this paper been cited?
0 times, according to OpenAlex on 19 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.
Is this paper free to read?
Yes. It is open access at arXiv (Cornell University): sign in and open the PDF from this page.
Who wrote this paper?
Xuanzhao Dong, Wenhui Zhu, Peijie Qiu, Xiwen Chen, Xiaobing Yu, Xin Li, Zhipeng Wang, Shao Tang, Gen Li, Yujian Xiong, Hao Wang, Yanxi Chen and 2 more.
What is the DOI of this paper?
10.48550/arxiv.2605.27960, permanently at https://doi.org/10.48550/arxiv.2605.27960. Sign in and open it from this page.
Bibliographic data from OpenAlex (CC0), last checked 19 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.