Can MLLMs Reason Beyond Language? VisReason: A Comprehensive Benchmark for Vision-Centric Reasoning
On the map: LLM reasoning → Multimodal reasoning
Abstract (arXiv)
Recent multimodal large language models (MLLMs) achieve strong performance on visual reasoning benchmarks, yet it remains unclear to what extent such performance reflects reasoning directly grounded in visual evidence. We introduce VisReason, a benchmark for vision-centric reasoning in everyday scenarios where perception and inference are tightly coupled. VisReason contains 1,505 questions across 10 categories spanning perceptual, structural, and conceptual reasoning. Our evaluation shows that VisReason poses a qualitatively different challenge from existing benchmarks, exposing substantial gaps between humans and current MLLMs and revealing limited benefits from test-time reasoning strategies. VisReason offers a focused diagnostic for evaluating vision-centric reasoning beyond language.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Multimodal reasoning, one of 17 directions in LLM reasoning. Builds on 11 of its 19 references that are in the map. First posted on arXiv 25 May 2026.
Builds on
11 of 19 references.
Closest ancestors · 2
Same direction, newest first
- 2025MiMo-VL Technical Report4 Jun
- 2025VGRP-Bench: Visual Grid Reasoning Puzzle Benchmark for Large Vision-Language Models29 Mar
Other references · 9
- 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
- 2025Qwen2.5-VL Technical Reportcited by 5,587
- 2024GPT-4o System Cardcited by 4,588
- 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,234
- 2025Qwen3-VL Technical Reportcited by 2,793
- 2025InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Modelscited by 1,834
- 2023Gemini: A Family of Highly Capable Multimodal Modelscited by 872
- 2025Kwai Keye-VL 1.5 Technical Reportcited by 73
- 2025Sudoku-Bench: Evaluating creative reasoning with Sudoku variantscited by 22
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2025VisuLogic: A Benchmark for Evaluating Visual Reasoning in Multi-modal Large Language Models
- 2024ProReason: Multi-Modal Proactive Reasoning with Decoupled Eyesight and Wisdom
- 2025VisRes Bench: On Evaluating the Visual Reasoning Capabilities of VLMs
- 2025VisReason: A Large-Scale Dataset for Visual Chain-of-Thought Reasoning
- 2026Do Vision-Language Models Truly Perform Vision Reasoning? A Rigorous Study of the Modality Gap
- 2025VERIFY: A Benchmark of Visual Explanation and Reasoning for Investigating Multimodal Reasoning Fidelity
- 2026MM-OPD: Towards One More Bottleneck Between Perception and Reasoning
- 2026VRIQ: Benchmarking and Analyzing Visual-Reasoning IQ of VLMs