Be Consistent! Enhancing Robust Visual Reasoning in LVLMs with Consistency Constraints
Large Vision-Language Models (LVLMs) struggle with robust visual reasoning. Existing benchmarks focus on symbolic math or simple vision tasks, lacking assessment of complex reasoning and logical consistency. This paper introduces ConVBench, a benchmark with logically equivalent question pairs across six categories, and two metrics: logical consistency and robust accuracy. The authors also propose ConVLM, using GRPO-based reinforcement learning with a novel consistency reward to improve LVLM reasoning.
-->
[Submitted on 23 Jul 2026]
Title:Be Consistent! Enhancing Robust Visual Reasoning in LVLMs with Consistency Constraints
View a PDF of the paper titled Be Consistent! Enhancing Robust Visual Reasoning in LVLMs with Consistency Constraints, by Liqiang Jing and 5 other authors
View PDF HTML (experimental)
Abstract:While Large Vision-Language Models (LVLMs) exhibit strong perceptual capabilities, they remain vulnerable in visual reasoning tasks. Existing benchmarks largely focus on symbolic mathematical or scientific problems and simple vision-centric tasks, offering limited assessment of complex visual reasoning and logical consistency, a critical requirement for reliable reasoning systems. We introduce ConVBench, a complex vision-centric reasoning benchmark in which each image is paired with two logically equivalent questions across six categories: action and state, complex counting, spatial reasoning, causal and intent understanding, commonsense reasoning, and temporal perception. To complement this benchmark, we define two evaluation metrics, logical consistency and robust accuracy, that jointly assess both the correctness and consistency of model responses. We further present ConVLM, which improves LVLM reasoning through Group Relative Policy Optimization (GRPO)-based reinforcement learning with a novel consistency reward. This method leverages automatically generated logically equivalent question-answer pairs and a dual-reward design combining accuracy- and consistency-based signals, encouraging agreement between paired responses. The framework functions effectively with or without strict answer supervision.
Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Cite as: arXiv:2607.21722 [cs.CV]
(or arXiv:2607.21722v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2607.21722
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Liqiang Jing [view email] [v1] Thu, 23 Jul 2026 18:04:51 UTC (3,456 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled Be Consistent! Enhancing Robust Visual Reasoning in LVLMs with Consistency Constraints, by Liqiang Jing and 5 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.CV
new | recent | 2026-07
Change to browse by:
cs
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
Loading...
Data provided by:
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
Author
Venue
Institution
Topic
About arXivLabs
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)