Pith. sign in

REVIEW 8 cited by

Simple o3: Towards Interleaved Vision-Language Reasoning

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2508.12109 v1 pith:PIZHHPHO submitted 2025-08-16 cs.CV cs.AI

Ye Wang , Qianglong Chen , Zejun Li , Siyuan Wang , Shijie Guo , Zhirui Zhang , Zhongyu Wei This is my paper
classification cs.CVcs.AI
keywords reasoninginterleavedvision-languagevisualmodelsimplecapabilitiesimage
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Multimodal Large Language Models (MLLMs) have shown impressive performance on vision-language tasks, but their long Chain-of-Thought (CoT) capabilities in multimodal scenarios remain underexplored. Inspired by OpenAI's o3 model, which emulates human-like ''thinking with image'' through iterative visual transformations and linguistic reasoning, we propose Simple o3, an end-to-end framework that integrates dynamic tool interactions (e.g., cropping, zooming, and reusing) into interleaved vision-language reasoning via supervised fine-tuning (SFT). Our approach features a scalable data synthesis pipeline that generates high-quality interleaved vision-language reasoning chains via an ''observe-reason-act'' cycle, complete with executable visual operations and rigorous verification, yielding the open-source TWI-Tools-146K dataset. Experimental results demonstrate Simple o3's superior performance on diverse benchmarks, outperforming existing approaches. By combining enhanced reasoning capabilities, Simple o3 establishes a powerful yet computationally affordable paradigm for advancing multimodal reasoning. Remarkably, we provide the first in-depth analysis of different interleaved reasoning strategies, offering insights into their impact on model performance. We found that by introducing additional visual tokens for interleaved vision-language reasoning, reusing and magnifying the original image significantly improves the model's visual reasoning and fine-grained perception, while image cropping based on precise visual grounding allows the model to effectively focus on key entities or regions, further enhancing its capabilities.

Discussion (0). Sign in to comment.

Forward citations

Cited by 8 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Diversity Over Frequency: Rethinking Tool Use in Visual Chain-of-Thought Agents

    cs.CV 2026-05 unverdicted novelty 7.0 of 10

    Visual CoT agents exhibit tool-use collapse where tool usage declines but task accuracy rises, and adding entropy regularization for rollout diversity produces the strongest performance.

  2. VideoThinker: Building Agentic VideoLLMs with LLM-Guided Tool Reasoning

    cs.CV 2026-01 unverdicted novelty 7.0 of 10

    VideoThinker uses LLM-generated synthetic tool trajectories in caption space grounded to video frames to train agentic VideoLLMs that outperform baselines on long-video benchmarks.

  3. Training Multi-Image Vision Agents via End2End Reinforcement Learning

    cs.CV 2025-12 unverdicted novelty 7.0 of 10

    IMAgent trains a multi-image vision agent via pure end-to-end RL with visual reflection tools and a two-layer motion trajectory masking strategy, reaching SOTA on single- and multi-image benchmarks while revealing too...

  4. BVS: Bayesian Visual Search with Multimodal Large Language Model for Fine-grained Perception

    cs.CV 2026-07 conditional novelty 6.0 of 10

    BVS combines early-stop attention rollout priors with a scale-aware non-stationary kernel and GP-UCB to locate tiny objects in UHR images more accurately and with fewer MLLM queries than prior visual-search methods.

  5. ROVER: Routing Object-Centric Visual Evidence for Grounded Multi-Image Reasoning

    cs.CV 2026-05 unverdicted novelty 6.0 of 10

    ROVER introduces a learnable routing plugin for object-centric visual evidence in MLLMs via token triplets and differential attention, reporting gains on MM-GCoT and VideoEspresso when integrated into Qwen2.5-VL-7B.

  6. Visual Reasoning through Tool-supervised Reinforcement Learning

    cs.CV 2026-04 unverdicted novelty 6.0 of 10

    ToolsRL trains MLLMs via a tool-specific then accuracy-focused RL curriculum to master visual tools for complex reasoning tasks.

  7. MAG-3D: Multi-Agent Grounded Reasoning for 3D Understanding

    cs.CV 2026-04 unverdicted novelty 5.0 of 10

    MAG-3D is a training-free multi-agent framework that coordinates planning, grounding, and coding agents with off-the-shelf VLMs to achieve grounded 3D reasoning and state-of-the-art benchmark results.

  8. Reinforced Visual Perception with Tools

    cs.CV 2025-09 conditional novelty 5.0 of 10

    ReVPT uses GRPO reinforcement learning with a cold-start SFT phase to make Qwen2.5-VL models call visual tools, improving perception benchmarks over SFT and text-only RL baselines.

Pith tools