Paper · 2026 · Open access
Mitigating Visual Context Degradation in Large Multimodal Models: A Training-Free Decoupled Agentic Framework
2026 IEEE/CVF Conference on Computer Vision and Pattern Recognition- FINDINGS Track (CVPRF), 9575–9586 · Published 1 June 2026 · DOI 10.1109/cvprf71909.2026.00917
In brief
“Mitigating Visual Context Degradation in Large Multimodal Models: A Training-Free Decoupled Agentic Framework” is a 2026 paper by Hongrui Jia and 3 co-authors, published in 2026 IEEE/CVF Conference on Computer Vision and Pattern Recognition- FINDINGS Track (CVPRF). OpenAlex counts 2 citations as of 19 November 2026, which puts it in the top 6% of works published in 2026 by citations. It is open access under a CC BY licence: the full text is free to read.
- Citations
- 2
- Most cited in
- 2026 (2)
- Field-weighted impact
- 2.59
- References
- 49
- Authors
- 4
Abstract
With the continuous expansion of Large Language Models (LLMs) and advances in reinforcement learning, LLMs have demonstrated exceptional reasoning capabilities, enabling them to address a wide range of complex problems. Inspired by these achievements, researchers have extended related techniques to Large Multimodal Models (LMMs). However, a critical limitation has emerged, reflected in the progressive loss of visual grounding. As the reasoning chain grows longer, LMMs tend to rely increasingly on the textual information generated in earlier steps, while the initially extracted visual information is rarely revisited or incorporated. This phenomenon often causes the reasoning process to drift away from the actual image content, resulting in visually implausible or even erroneous conclusions. To overcome this fundamental limitation, we propose a novel, trainingfree agentic paradigm that Decouples cognitive Reasoning from visual Perception (DRP). In this framework, a powerful LLM serves as a strategic Reasoner, orchestrating the inference process by explicitly querying an LMM, acting as a dedicated Observer, to retrieve fine-grained visual details on demand. This approach is lightweight, model-agnostic, and plug-and-play, necessitating no additional training or architectural modifications. Extensive experiments demonstrate our framework DRP's efficacy in regulating the visual reasoning trajectory, significantly mitigating reasoning drift, and enforcing robust visual grounding. Notably, on the MathVision benchmark, the integration of Qwen2.5-VL-7B and Qwen3-32B achieves an accuracy of 47.2%, outperforming GPT-4o's 40.6%. These findings underscore the potential of our approach to enhance multimodal reasoning reliability without the need for costly retraining. Our code is publicly available at https://github.com/hongruijia/DRP.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Jia, H., Jiang, C., Zhang, S., & Ye, W. (2026). Mitigating Visual Context Degradation in Large Multimodal Models: A Training-Free Decoupled Agentic Framework. 2026 IEEE/CVF Conference on Computer Vision and Pattern Recognition- FINDINGS Track (CVPRF), 9575–9586. https://doi.org/10.1109/cvprf71909.2026.00917
MLA 9
Jia, Hongrui, et al. “Mitigating Visual Context Degradation in Large Multimodal Models: A Training-Free Decoupled Agentic Framework.” 2026 IEEE/CVF Conference on Computer Vision and Pattern Recognition- FINDINGS Track (CVPRF), 2026, pp. 9575–9586. https://doi.org/10.1109/cvprf71909.2026.00917.
Chicago
Jia, Hongrui, Chaoya Jiang, Shikun Zhang, and Wei Ye. 2026. “Mitigating Visual Context Degradation in Large Multimodal Models: A Training-Free Decoupled Agentic Framework.” 2026 IEEE/CVF Conference on Computer Vision and Pattern Recognition- FINDINGS Track (CVPRF): 9575–9586. https://doi.org/10.1109/cvprf71909.2026.00917.
Harvard
Jia, H. et al. (2026) ‘Mitigating Visual Context Degradation in Large Multimodal Models: A Training-Free Decoupled Agentic Framework’, 2026 IEEE/CVF Conference on Computer Vision and Pattern Recognition- FINDINGS Track (CVPRF), pp. 9575–9586. doi:10.1109/cvprf71909.2026.00917.
IEEE
H. Jia, C. Jiang, S. Zhang, and W. Ye, “Mitigating Visual Context Degradation in Large Multimodal Models: A Training-Free Decoupled Agentic Framework,” 2026 IEEE/CVF Conference on Computer Vision and Pattern Recognition- FINDINGS Track (CVPRF), pp. 9575–9586, 2026, doi: 10.1109/cvprf71909.2026.00917.
BibTeX
@article{jia2026mitigating,
title = {{Mitigating Visual Context Degradation in Large Multimodal Models: A Training-Free Decoupled Agentic Framework}},
author = {Jia, Hongrui and Jiang, Chaoya and Zhang, Shikun and Ye, Wei},
journal = {2026 IEEE/CVF Conference on Computer Vision and Pattern Recognition- FINDINGS Track (CVPRF)},
year = {2026},
pages = {9575--9586},
doi = {10.1109/cvprf71909.2026.00917},
url = {https://doi.org/10.1109/cvprf71909.2026.00917}
}
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
- large multimodal models
- visual grounding
- multimodal reasoning
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?
Jia, H., Jiang, C., Zhang, S., & Ye, W. (2026). Mitigating Visual Context Degradation in Large Multimodal Models: A Training-Free Decoupled Agentic Framework. 2026 IEEE/CVF Conference on Computer Vision and Pattern Recognition- FINDINGS Track (CVPRF), 9575–9586. https://doi.org/10.1109/cvprf71909.2026.00917
How many times has this paper been cited?
2 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: sign in and open the full text from this page.
Who wrote this paper?
Hongrui Jia, Chaoya Jiang, Shikun Zhang, Wei Ye.
What is the DOI of this paper?
10.1109/cvprf71909.2026.00917, permanently at https://doi.org/10.1109/cvprf71909.2026.00917. 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.