cs.LGNov 14, 2026

LogT: Logically Think with Images for Visual Search

Authors: Yanjun Fu, Quanzeng You, Jiadong Guo, Yujie Lu, Sai Charitha Akula, Haiping Wu, Miao Liu, Sanghamitra Dutta, 

Organizations: Meta Superintelligence Labs (MSL) · University of Maryland

Abstract

Thinking with images empowers vision-language models (VLMs) to use an external Zoom-in tool for fine-grained visual search. Recent works employ supervised fine-tuning (SFT) and reinforcement learning (RL) with tool-specific designs to enable such capabilities. However, achieving logical reasoning alongside accurate and efficient tool use remains a significant challenge. To address this, we shift the focus away from increasing RL complexity and instead identify SFT data quality as a key bottleneck. We then propose LogT (short for Logically Think with Images), a fully automated SFT data engine that explicitly enforces logic-chain-guided reasoning. LogT generates an SFT dataset featuring challenging, visually grounded questions and structured tool-use traces with human-like logic. We validate our approach by fine-tuning two base models, Qwen2.5-VL-7B and InternVL3.5-8B, on the dataset and evaluating them across seven benchmark test sets. We show that LogT-SFT based on Qwen2.5-VL-7B outperforms the state-of-the-art trained with both SFT and RL, despite using 3.1×\times less data and >120×>120\times fewer GPU hours. Furthermore, applying a simple RL recipe without tool-specific design on LogT-SFT yields a model that surpasses prior methods with complex RL setups by a substantial 9.59-point margin. In addition to strong performance gains, our model exhibits higher tool-use accuracy and efficiency, as well as more structured and logically consistent reasoning traces, directly validating the effectiveness of our approach.

Figures & tables

Appendix figures & tables13 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. ToolVision: Learning When and How to Use Visual Tools with Capability-Aligned Supervision

    Aug 9, 2026Delin Mao, Chenghao Sun, Jingwei Song +2Tool Use in VLMsTool-Using Vision-Language Agents

  2. VisualSphinx: Large-Scale Synthetic Vision Logic Puzzles for RL

    May 29, 2025Yichen Feng, Zhangchen Xu, Fengqing Jiang +5Synthetic Data Generation for VLMsVisual Reasoning

  3. Distilling Visual Reasoning into Text Space

    Sep 28, 2026Wenhan Yang, Nilay Naharas, Ali Payani +1CoT DistillationVLM Distillation