Optical Reasoning: Rethinking Images as an Expressive Reasoning Medium Beyond Text

• URL: http://arxiv.org/abs/2606.09585v1
• Date: Mon, 08 Jun 2026 14:58:59 GMT
• Title: Optical Reasoning: Rethinking Images as an Expressive Reasoning Medium Beyond Text
• Abstract summary: Chain-of-Thought (CoT) improves the performance of Large Language Models (LLMs) and has been extended to Multimodal Large Language Models (MLLMs)<n>We propose optical reasoning, which treats images as a standalone reasoning medium.
• Score: 15.504855816862587
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Chain-of-Thought (CoT) improves the performance of Large Language Models (LLMs) and has been extended to Multimodal Large Language Models (MLLMs). More recent work further moves from text-based multimodal reasoning toward interleaved-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 propose optical 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, and graphical-based optical reasoning, which composes text and graphical elements into structured visual rationales. Across mathematical, scientific, and interleaved-modal reasoning benchmarks, optical reasoning can 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 the token efficiency of text reasoning. These results show that images can effectively and efficiently encode rationales while providing a unified visual canvas for reasoning.

Related papers

• Look on Demand: A Cognitive Scheduling Framework for Visual Evidence Acquisition in Multimodal Reasoning [82.86343313807158]
  We argue that a central challenge is how and when visual evidence is introduced into the reasoning process.<n>Motivated by this insight, we propose CSMR, a multimodal reasoning framework in which a language model controls the reasoning process.
  arXiv  Detail & Related papers  (2026-05-27T08:43:13Z)
• UniVLR: Unifying Text and Vision in Visual Latent Reasoning for Multimodal LLMs [36.74593429466411]
  We propose a unified visual latent reasoning framework that treats textual reasoning and auxiliary visual evidence as a shared visual workspace.<n>We show that UniVLR outperforms prior visual latent reasoning methods while using substantially fewer generated reasoning tokens.
  arXiv  Detail & Related papers  (2026-05-12T09:40:03Z)
• Walk the Talk: Bridging the Reasoning-Action Gap for Thinking with Images via Multimodal Agentic Policy Optimization [89.68681087743876]
  Multimodal Large Language Models (MLLMs) have incentivized models to think with images'' by actively invoking visual tools during multi-turn reasoning.<n>The common Reinforcement Learning practice of relying on outcome-based rewards ignores the fact that textual plausibility often masks executive failure.<n>We introduce Multimodal Agentic Policy Optimization (MAPO), bridging the gap between textual reasoning and visual actions generated by models within their Multimodal Chain-of-Thought.
  arXiv  Detail & Related papers  (2026-04-08T07:48:07Z)
• Thinking with Comics: Enhancing Multimodal Reasoning through Structured Visual Storytelling [25.305621689469444]
  Thinking with Comics is a visual reasoning paradigm that uses comics as a high information-density medium positioned between images and videos.<n>Comics preserve temporal structure, embedded text, and narrative coherence while requiring significantly lower reasoning cost.
  arXiv  Detail & Related papers  (2026-02-02T18:43:57Z)
• Mull-Tokens: Modality-Agnostic Latent Thinking [83.76982012426055]
  Mull-Tokens are modality-agnostic latent tokens pre-trained to hold intermediate information in either image or text modalities.<n>We demonstrate that Mull-Tokens improve upon several baselines utilizing text-only reasoning or interleaved image-text reasoning.
  arXiv  Detail & Related papers  (2025-12-11T18:59:08Z)
• Monet: Reasoning in Latent Visual Space Beyond Images and Language [55.424507246294326]
  "Thinking with images" has emerged as an effective paradigm for advancing visual reasoning.<n>Existing methods fall short of human-like abstract visual thinking.<n>We introduce Monet, a training framework that enables multimodal large language models to reason directly within the latent visual space.
  arXiv  Detail & Related papers  (2025-11-26T13:46:39Z)
• ThinkMorph: Emergent Properties in Multimodal Interleaved Chain-of-Thought Reasoning [76.95203056566191]
  Multimodal reasoning requires iterative coordination between language and vision, yet it remains unclear what constitutes a meaningful interleaved chain of thought.<n>We build ThinkMorph, a unified model fine-tuned on approximately 24K high-quality interleaved reasoning traces spanning tasks with varying visual engagement.<n>ThinkMorph learns to generate progressive text-image reasoning steps that concretely manipulate visual content while maintaining coherent verbal logic.
  arXiv  Detail & Related papers  (2025-10-30T17:51:38Z)
• Visualizing Thought: Conceptual Diagrams Enable Robust Planning in LMMs [59.66595230543127]
  Conceptual diagrams externalize mental models, abstracting irrelevant details to efficiently capture how entities interact.<n>Large Language Models (LLMs) and Large MultiModal Models (LMMs) predominantly reason through text.<n>We propose Visual Thinking, a generalizable framework that enables LMMs to reason through multiple chains of self-generated conceptual diagrams.
  arXiv  Detail & Related papers  (2025-03-14T18:27:02Z)
• A Neural Divide-and-Conquer Reasoning Framework for Image Retrieval from Linguistically Complex Text [23.854023255928208]
  We propose an end-to-end Neural Divide-and-Conquer Reasoning framework, dubbed NDCR. It contains three main components: 1) Divide: a proposition generator divides the compound proposition text into simple proposition sentences and produces their corresponding representations, 2) Conquer: a pretrained visual-linguistic interactor achieves the interaction between proposition sentences and images, and 3) Combine: a neural-symbolic reasoner combines the above reasoning states to obtain the final solution.
  arXiv  Detail & Related papers  (2023-05-03T16:55:00Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.