DeepSketcher: Internalizing Visual Manipulation for Multimodal Reasoning

• URL: http://arxiv.org/abs/2509.25866v1
• Date: Tue, 30 Sep 2025 07:02:01 GMT
• Title: DeepSketcher: Internalizing Visual Manipulation for Multimodal Reasoning
• Authors: Chi Zhang, Haibo Qiu, Qiming Zhang, Zhixiong Zeng, Lin Ma, Jing Zhang,
• Abstract summary: "Thinking with images" is a shift in Vision Language Models from text-dominant chain-of-thought to image-interactive reasoning.<n>We present DeepSketcher, a comprehensive suite comprising both an image-text interleaved dataset and a self-contained model.<n>We design a model that performs interleaved image-text reasoning and generates "visual thoughts" by operating directly in the visual embedding space.
• Score: 16.880791276029964
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The "thinking with images" paradigm represents a pivotal shift in the reasoning of Vision Language Models (VLMs), moving from text-dominant chain-of-thought to image-interactive reasoning. By invoking visual tools or generating intermediate visual representations, VLMs can iteratively attend to fine-grained regions, enabling deeper image understanding and more faithful multimodal reasoning. As an emerging paradigm, however, it still leaves substantial room for exploration in data construction accuracy, structural design, and broader application scenarios, which offer rich opportunities for advancing multimodal reasoning. To further advance this line of work, we present DeepSketcher, a comprehensive suite comprising both an image-text interleaved dataset and a self-contained model. The dataset contains 31k chain-of-thought (CoT) reasoning trajectories with diverse tool calls and resulting edited images, covering a wide range of data types and manipulation instructions with high annotation accuracy. Building on this resource, we design a model that performs interleaved image-text reasoning and natively generates "visual thoughts" by operating directly in the visual embedding space, rather than invoking external tools and repeatedly re-encoding generated images. This design enables tool-free and more flexible "thinking with images". Extensive experiments on multimodal reasoning benchmarks demonstrate strong performance, validating both the utility of the dataset and the effectiveness of the model design.

Related papers

• Chatting with Images for Introspective Visual Thinking [50.7747647794877]
  ''Chatting with images'' is a new framework that reframes visual manipulation as language-guided feature modulation.<n>Under the guidance of expressive language prompts, the model dynamically performs joint re-encoding over multiple image regions.<n>ViLaVT achieves strong and consistent improvements on complex multi-image and video-based spatial reasoning tasks.
  arXiv  Detail & Related papers  (2026-02-11T17:42:37Z)
• DeepImageSearch: Benchmarking Multimodal Agents for Context-Aware Image Retrieval in Visual Histories [52.57197752244638]
  We introduce DeepImageSearch, a novel agentic paradigm that reformulates image retrieval as an autonomous exploration task.<n>Models must plan and perform multi-step reasoning over raw visual histories to locate targets based on implicit contextual cues.<n>We construct DISBench, a challenging benchmark built on interconnected visual data.
  arXiv  Detail & Related papers  (2026-02-11T12:51:10Z)
• Deep But Reliable: Advancing Multi-turn Reasoning for Thinking with Images [53.373427633330515]
  We propose DRIM, a model that enables deep but reliable multi-turn reasoning when thinking with images in its multimodal CoT.<n>Based on a high-resolution image dataset, we construct high-difficulty and verifiable visual question-answer pairs.<n>In the SFT stage, we collect tool trajectories as cold-start data, guiding a multi-turn reasoning pattern.<n>In the RL stage, we introduce redundancy-penalized policy optimization, which incentivizes the model to develop a self-reflective reasoning pattern.
  arXiv  Detail & Related papers  (2025-12-19T07:44:43Z)
• Visual Bridge: Universal Visual Perception Representations Generating [27.034175361589572]
  We propose a universal visual perception framework based on flow matching that can generate diverse visual representations across multiple tasks.<n>Our approach formulates the process as a universal flow-matching problem from image patch tokens to task-specific representations.<n>Our model achieves competitive performance in both zero-shot and fine-tuned settings, outperforming prior generalist and several specialist models.
  arXiv  Detail & Related papers  (2025-11-11T06:25:30Z)
• Simple o3: Towards Interleaved Vision-Language Reasoning [38.46230601239066]
  We propose Simple o3, an end-to-end framework that integrates dynamic tool interactions into interleaved vision-language reasoning.<n>Our approach features a scalable data synthesis pipeline that generates high-quality interleaved vision-language reasoning chains.<n> Experimental results demonstrate Simple o3's superior performance on diverse benchmarks, outperforming existing approaches.
  arXiv  Detail & Related papers  (2025-08-16T17:15:39Z)
• Weaving Context Across Images: Improving Vision-Language Models through Focus-Centric Visual Chains [31.828341309787042]
  Vision-language models (VLMs) achieve remarkable success in single-image tasks.<n>Real-world scenarios often involve intricate multi-image inputs, leading to a notable performance decline.<n>We propose Focus-Centric Visual Chain, a novel paradigm that enhances VLMs'perception, comprehension, and reasoning abilities in multi-image scenarios.
  arXiv  Detail & Related papers  (2025-04-28T19:02:18Z)
• Towards Text-Image Interleaved Retrieval [49.96332254241075]
  We introduce the text-image interleaved retrieval (TIIR) task, where the query and document are interleaved text-image sequences.<n>We construct a TIIR benchmark based on naturally interleaved wikiHow tutorials, where a specific pipeline is designed to generate interleaved queries.<n>We propose a novel Matryoshka Multimodal Embedder (MME), which compresses the number of visual tokens at different granularity.
  arXiv  Detail & Related papers  (2025-02-18T12:00:47Z)
• MaVEn: An Effective Multi-granularity Hybrid Visual Encoding Framework for Multimodal Large Language Model [49.931663904599205]
  MaVEn is an innovative framework designed to enhance the capabilities of Multimodal Large Language Models (MLLMs) in multi-image reasoning. We show that MaVEn significantly enhances MLLMs' understanding in complex multi-image scenarios, while also improving performance in single-image contexts.
  arXiv  Detail & Related papers  (2024-08-22T11:57:16Z)
• From Redundancy to Relevance: Information Flow in LVLMs Across Reasoning Tasks [33.476693301050275]
  We conduct experiments with truncation strategies across various LVLMs for visual question answering and image captioning tasks. By exploring the information flow from the perspective of visual representation contribution, we observe that it tends to converge in shallow layers but diversify in deeper layers.
  arXiv  Detail & Related papers  (2024-06-04T13:52:54Z)
• 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)
• Multi-modal Auto-regressive Modeling via Visual Words [96.25078866446053]
  We propose the concept of visual tokens, which maps the visual features to probability distributions over Large Multi-modal Models' vocabulary. We further explore the distribution of visual features in the semantic space within LMM and the possibility of using text embeddings to represent visual information.
  arXiv  Detail & Related papers  (2024-03-12T14:58:52Z)
• StableLLaVA: Enhanced Visual Instruction Tuning with Synthesized Image-Dialogue Data [129.92449761766025]
  We propose a novel data collection methodology that synchronously synthesizes images and dialogues for visual instruction tuning. This approach harnesses the power of generative models, marrying the abilities of ChatGPT and text-to-image generative models. Our research includes comprehensive experiments conducted on various datasets.
  arXiv  Detail & Related papers  (2023-08-20T12:43:52Z)

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.