Optical Reasoning: Rethinking Images as an Expressive Reasoning Medium Beyond Text
Summary
Proposes optical reasoning, using images as a standalone reasoning medium for language and multimodal tasks, achieving higher token efficiency than traditional text-based approaches.
View Cached Full Text
Cached at: 06/09/26, 08:40 AM
Paper page - Optical Reasoning: Rethinking Images as an Expressive Reasoning Medium Beyond Text
Source: https://huggingface.co/papers/2606.09585
Abstract
Optical reasoning uses images as a standalone reasoning medium for language and multimodal tasks, achieving higher token efficiency than traditional text-based approaches.
Chain-of-Thought(CoT) improves the performance ofLarge Language Models(LLMs) and has been extended toMultimodal Large Language Models(MLLMs). More recent work further moves from text-based multimodal reasoning towardinterleaved-modal reasoning, where intermediate steps can incorporate both textual rationales and visual evidence. In this work, we propose a bolder and more ambitious idea: could images alone serve as the reasoning medium for both language and multimodal tasks? To explore this, we proposeoptical reasoning, which treats images as a standalone reasoning medium. We instantiate this concept with two variants:typographic-based optical reasoning, which optimizes visual layouts for compact rationale rendering, andgraphical-based optical reasoning, which composes text and graphical elements into structuredvisual rationales. Across mathematical, scientific, andinterleaved-modal reasoningbenchmarks,optical reasoningcan match or even exceed traditional text reasoning while reducing reasoning tokens by an average of 28.57% on language tasks and 16% on multimodal tasks, achieving 1.96 times thetoken efficiencyof text reasoning. These results show that images can effectively and efficiently encode rationales while providing a unified visual canvas for reasoning.
View arXiv pageView PDFGitHub0Add to collection
Get this paper in your agent:
hf papers read 2606\.09585
Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2606.09585 in a model README.md to link it from this page.
Datasets citing this paper1
#### ModalityDance/Optical-Reasoning-4k Viewer• Updated25 minutes ago • 9.74k • 1
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2606.09585 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
Look Light, Think Heavy: What Multimodal Chain-of-Thought Reasoning Can and Cannot Do
This paper systematically evaluates multimodal Chain-of-Thought reasoning across 12 tasks, finding it selectively effective for reasoning tasks but detrimental for perception tasks, and identifies a 'Look Light, Think Heavy' pattern where visual introspection declines during reasoning.
Perceive-to-Reason: Decoupling Perception and Reasoning for Fine-Grained Visual Reasoning
Introduces Perceive-to-Reason (P2R), a framework that decouples visual perception from reasoning in vision-language models using a two-stage process and a role-aware reinforcement learning strategy, achieving state-of-the-art results on fine-grained visual reasoning benchmarks.
ETCHR: Editing To Clarify and Harness Reasoning
ETCHR is a novel image editing approach that decouples visual reasoning from image generation, using a two-stage training process (Reasoning Imitation and Reasoning Enhancement) to improve multimodal language model performance across five visual reasoning tasks. It achieves consistent gains of 4-5% Pass@1 on models like Qwen3-VL-8B, Gemini-3.1-Flash-Lite, and Kimi K2.5.
Thinking with images
OpenAI releases o3 and o4-mini models that can reason with images in their chain-of-thought process, enabling visual understanding through native image manipulation tools like cropping and zooming without separate specialized models. These models achieve state-of-the-art performance on multimodal benchmarks including STEM questions, chart reading, and visual search tasks.
Bad Seeing or Bad Thinking? Rewarding Perception for Vision-Language Reasoning
This paper introduces a reinforcement learning framework that improves perception-reasoning synergy in vision-language models by explicitly rewarding perceptual fidelity, using a 'blindfolded reasoning' proxy and structured verbal verification to address ambiguity in modality credit assignment.