Sep 2026· Conference on Empirical Methods in Natural Language Processing· Vol abs/2609.14779, pp. 40616-40632· 1 citation· 34 references
Computer Science
TL;DR
Func-R1 is proposed, which synergistically harmonizes precise visual perception and rigorous logical reasoning and delivers the optimal performance among open-source MLLMs, even surpassing GPT-5 with an 8.4% improvement on MathVerse's function-oriented tasks.
Abstract
Performing deliberate mathematical reasoning in visual contexts is a hallmark of advanced Multimodal Large Language Models (MLLMs) and requires a sophisticated synthesis of perceptual grounding and symbolic logic. However, in the realm of mathematical functions, our investigation reveals a critical modality interference phenomenon: even advanced models, while performing textual computational reasoning, tend to disregard or misinterpret essential visual cues. To address this challenge, we propose Func-R1, which synergistically harmonizes precise visual perception and rigorous logical reasoning. Concretely, built upon an explicitly decoupled architecture, we employ a hierarchical post-training framework to progressively identify critical visual evidence and conduct in-depth theoretical reasoning. Furthermore, the Perception-Aligned Theoretic Optimization (PATO) strategy is proposed to steer policy updating towards internalizing fundamental theoretical properties while dynamically rectifying heterogeneous visual information throughout the reasoning process. Extensive experiments across diverse benchmarks demonstrate that Func-R1 delivers the optimal performance among open-source MLLMs, even surpassing GPT-5 with an 8.4% improvement on MathVerse's function-oriented tasks.
The introduction of long chain-of-thought (CoT) reasoning and external tools has substantially improved the performance of multimodal large language models (MLLMs) on general and reasoning tasks. However, existing fast-slow reasoning paradigms mainly regulate textual reasoning length, while visual computation introduce...
Quan-Lin Chen, Xiu-Wei Chen, Kun Xiang et al.· 0 citations
Multimodal Large Language Models (MLLMs) have shown strong performance on mathematical reasoning tasks that combine textual and visual information. However, final-answer accuracy offers limited insight into whether errors arise from incorrect perception, modality-dependent information extraction, or downstream reasonin...
Aniket Mitra, Arpit Chhabra, Aditya Desai et al.· Proceedings of the 35th ACM...· 0 citations
Recent multimodal large language models achieve promising results on visual reasoning tasks, approaching human-level performance on existing STEM benchmarks. However, these capabilities degrade under complex real-world scenes. We identify that chain-of-thought reasoning remains largely confined to the linguistic space,...
Fan Yang, Li-Feng Chen, Yi-Pu Wang et al.· Proceedings of the 2026 Conf...· 0 citations
MM-OPD is introduced, a multimodal on-policy self-distillation framework for symbolic-to-visual correction that transfers guidance from symbolic-conditioned behavior to the image-conditioned policy through residual token-level targets, steering the model toward correct visual evidence.
This work proposes Dense Reasoning Trace, a paradigm that departs from natural-language-centered CoT by expressing reasoning as compact structured traces, which include concise intermediate states with symbolic connectors and disentangle visual observations from logical deductions, and proposes the Trace-Grounded Reinf...
Wan Xu, Yuan-Fan Guo, Kevin Han et al.· arXiv.org· 0 citations
Extended reasoning has substantially improved multimodal large language models (MLLMs) on tasks requiring compositional and relational inference. However, multimodal grounding imposes a distinct requirement: reasoning must ultimately converge on a target-specific compact representation that supports precise localizatio...
J. Seo, Hyeongjun Kwon, Jiyoung Lee et al.· 0 citations