Reasoning Within the Mind: Dynamic Multimodal Interleaving in Latent Space
- URL: http://arxiv.org/abs/2512.12623v2
- Date: Wed, 17 Dec 2025 21:01:39 GMT
- Title: Reasoning Within the Mind: Dynamic Multimodal Interleaving in Latent Space
- Authors: Chengzhi Liu, Yuzhe Yang, Yue Fan, Qingyue Wei, Sheng Liu, Xin Eric Wang,
- Abstract summary: We propose a test-time Dynamic Multimodal Latent Reasoning framework.<n>It employs confidence-guided latent policy gradient optimization to latent think tokens for in-depth reasoning.<n> Experiments across seven multimodal reasoning benchmarks and various model architectures demonstrate that DMLR significantly improves reasoning and perception performance.
- Score: 46.05748768260013
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Recent advancements in Multimodal Large Language Models (MLLMs) have significantly enhanced cross-modal understanding and reasoning by incorporating Chain-of-Thought (CoT) reasoning in the semantic space. Building upon this, recent studies extend the CoT mechanism to the visual modality, enabling models to integrate visual information during reasoning through external tools or explicit image generation. However, these methods remain dependent on explicit step-by-step reasoning, unstable perception-reasoning interaction and notable computational overhead. Inspired by human cognition, we posit that thinking unfolds not linearly but through the dynamic interleaving of reasoning and perception within the mind. Motivated by this perspective, we propose DMLR, a test-time Dynamic Multimodal Latent Reasoning framework that employs confidence-guided latent policy gradient optimization to refine latent think tokens for in-depth reasoning. Furthermore, a Dynamic Visual Injection Strategy is introduced, which retrieves the most relevant visual features at each latent think token and updates the set of best visual patches. The updated patches are then injected into latent think token to achieve dynamic visual-textual interleaving. Experiments across seven multimodal reasoning benchmarks and various model architectures demonstrate that DMLR significantly improves reasoning and perception performance while maintaining high inference efficiency.
Related papers
- Multimodal Latent Reasoning via Hierarchical Visual Cues Injection [16.779425236020433]
This work posits that robust reasoning should evolve within a latent space, integrating multimodal signals seamlessly.<n>We propose a novel framework that instills deliberate, "slow thinking" without depending on superficial textual rationales.<n>We show that test-time scaling is effective when incorporating vision knowledge, and that integrating hierarchical information significantly enhances the model's understanding of complex scenes.
arXiv Detail & Related papers (2026-02-05T06:31:12Z) - Interleaved Latent Visual Reasoning with Selective Perceptual Modeling [42.93438443502933]
Interleaved reasoning paradigms enhance Multimodal Large Language Models (MLLMs) with visual feedback but are hindered by the prohibitive computational cost.<n>A promising alternative, latent visual reasoning, circumvents this bottleneck yet currently forces a critical trade-off.<n>We introduce Interleaved Latent Visual Reasoning (ILVR), a framework that unifies dynamic state evolution with precise perceptual modeling.
arXiv Detail & Related papers (2025-12-05T12:09:39Z) - 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) - Reinforcing Spatial Reasoning in Vision-Language Models with Interwoven Thinking and Visual Drawing [62.447497430479174]
Drawing to reason in space is a novel paradigm that enables LVLMs to reason through elementary drawing operations in the visual space.<n>Our model, named VILASR, consistently outperforms existing methods across diverse spatial reasoning benchmarks.
arXiv Detail & Related papers (2025-06-11T17:41:50Z) - Imagine while Reasoning in Space: Multimodal Visualization-of-Thought [70.74453180101365]
Chain-of-Thought (CoT) prompting has proven highly effective for enhancing complex reasoning in Large Language Models (LLMs) and Multimodal Large Language Models (MLLMs)<n>We propose a new reasoning paradigm, Multimodal Visualization-of-Thought (MVoT)<n>It enables visual thinking in MLLMs by generating image visualizations of their reasoning traces.
arXiv Detail & Related papers (2025-01-13T18:23:57Z) - Cantor: Inspiring Multimodal Chain-of-Thought of MLLM [83.6663322930814]
We argue that converging visual context acquisition and logical reasoning is pivotal for tackling visual reasoning tasks.
We propose an innovative multimodal CoT framework, termed Cantor, characterized by a perception-decision architecture.
Our experiments demonstrate the efficacy of the proposed framework, showing significant improvements in multimodal CoT performance.
arXiv Detail & Related papers (2024-04-24T17:59:48Z)
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.