aiqbench analysis
This research addresses a critical gap in vision-language models by introducing MIRROR, a reinforcement learning method that leverages inconsistencies between text-only and visual inputs to improve geometric reasoning consistency. By constructing paired datasets where different modalities elicit distinct behaviors, the approach uses high-performing views as teachers to train others via reverse-KL optimization, resulting in more robust multimodal performance across benchmarks.
MIRROR: Learning from the Other View for Multi-Modal Reasoning
arXiv:2607.21552v1 Announce Type: new Abstract: Unlike large language models (LLMs) that exhibit strong reasoning capabilities, vision-language models (VLMs) struggle with visual reasoning, even on geometry problems that admit equivalent text, diagram, and combined diagram+text views. We show that these views often elicit different behaviors: a model may solve a problem from text but fail on the corresponding diagram, or succeed visually while failing textually. This inconsistency suggests that different views expose complementary reasoning paths and failure modes that standard multimodal post-training does not fully exploit. To study and exploit this phenomenon, we construct ODA-Data, a high-quality paired multimodal geometry dataset with text-dominant, image-dominant, and combined image+text views of the same problems, together with splits for training and evaluating modality-dependent reasoning behaviors. We then develop Modality-Informed Reciprocal Reasoning Optimization (MIRROR), a reinforcement learning approach for improving multimodal reasoning via self supervision. For each problem, MIRROR evaluates the model under all views, selects the best-performing view as a teacher, and trains other views with a reverse-KL objective towards the teacher. Across reasoning benchmarks that evaluate on geometry problems, MIRROR improves over standard RL and yields more accurate and consistent behavior across modalities
Abstract:Unlike large language models (LLMs) that exhibit strong reasoning capabilities, vision-language models (VLMs) struggle with visual reasoning, even on geometry problems that admit equivalent text, diagram, and combined diagram+text views. We show that these views often elicit different behaviors: a model may solve a problem from text but fail on the corresponding diagram, or succeed visually while failing textually. This inconsistency suggests that different views expose complementary reasoning paths and failure modes that standard multimodal post-training does not fully exploit. To study and exploit this phenomenon, we construct ODA-Data, a high-quality paired multimodal geometry dataset with text-dominant, image-dominant, and combined image+text views of the same problems, together with splits for training and evaluating modality-dependent reasoning behaviors. We then develop Modality-Informed Reciprocal Reasoning Optimization (MIRROR), a reinforcement learning approach for improving multimodal reasoning via self supervision. For each problem, MIRROR evaluates the model under all views, selects the best-performing view as a teacher, and trains other views with a reverse-KL objective towards the teacher. Across reasoning benchmarks that evaluate on geometry problems, MIRROR improves over standard RL and yields more accurate and consistent behavior across modalities
Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
arXiv:2607.21552 [cs.AI]
(or arXiv:2607.21552v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2607.21552 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
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.
Full text under the publisher's open licence · 282 words
Related coverage
Build a Physical AI model factory with NVIDIA Cosmos 3 on SageMaker HyperPod
AWS · Sep 4, 2026
Why do models task game?
AI Alignment Forum · Aug 6, 2026
Brief independent investigation of agents’ behavior, reasoning and collaboration in the OpenAI / Hugging Face hacking incident
METR · Aug 26, 2026
Biomedical Imaging's Real Bottleneck Is the Data, Not the Model
Databricks · Oct 8, 2026