paper-with-me

홈 › Papers

From Bounding Boxes to Visual Reasoning: An On-Policy Data Annotation Tool for Vision-Language Models

2026-06-17 · Like Zhang, Runliang Niu, Shiqi Wang, Xiyu Hu, Qianli Xing, Pan Wang, Qingzu He, Qi Wang arxiv

Vision-language models (VLMs) are rapidly advancing toward sophisticated grounded structured visual reasoning. Training models for such advanced capabilities demands a new genre of data that seamlessly unifies spatial coordinates, open-vocabulary descriptions, structured attributes, and topological relationships into a singular representation. However, existing data annotation tools fundamentally fail to meet these intricate demands, suffering from three systematic bottlenecks: limited expressiveness, severe annotation-training decoupling, and poor data reusability. To bridge this infrastructure gap, we introduce an open-source annotation tool, ScreenAnnotator. First, we define a unified annotation atom schema that binds spatial, semantic, and structural primitives into a single unit. Second, we implement an on-policy annotation loop embedded with a Bayesian Annotation Verifier (BAV). Finally, we design a template-driven multi-task data synthesis process dynamically transforms static atoms into diverse multi-dimensional reasoning tasks, eliminating redundant re-annotation. The on-policy loop drives the annotation accept rate to nearly 100% on flowcharts and 77% on GUI screenshots, while steadily reducing per-image annotation time as labeled data accumulate. In the flowchart scenario, fine-tuning a VLM yields 76.1% average accuracy, which is a 35.1% point absolute gain. Our code is available at: https://github.com/WnQinm/Annotator.

📄 PDF Abstract BibTeX arXiv:2606.18846

Code (0)

등록된 구현이 없습니다.

Tasks

Visual Reasoning

Similar Papers 제목 키워드 기반

Visual Attention Reasoning via Hierarchical Search and Self-Verification

2025-10-21 · Wei Cai, Jian Zhao, Yuchen Yuan, Tianle Zhang 외 arxiv

Multimodal Large Language Models (MLLMs) frequently hallucinate due to their reliance on fragile, linear reasoning and weak visual grounding. We propose Visual Attention Reasoning (VAR), a reinforcement learning framewor…

Reinforcement LearningVisual Grounding

Unsupervised Visual Chain-of-Thought Reasoning via Preference Optimization

2025-04-25 · Kesen Zhao, Beier Zhu, Qianru Sun, Hanwang Zhang

Chain-of-thought (CoT) reasoning greatly improves the interpretability and problem-solving abilities of multimodal large language models (MLLMs). However, existing approaches are focused on text CoT, limiting their abili…

Spatial Reasoning

EarthSpatialBench: Benchmarking Spatial Reasoning Capabilities of Multimodal LLMs on Earth Imagery

2026-02-17 · Zelin Xu, Yupu Zhang, Saugat Adhikari, Saiful Islam 외 arxiv

Benchmarking spatial reasoning in multimodal large language models (MLLMs) has attracted growing interest in computer vision due to its importance for embodied AI and other agentic systems that require precise interactio…

Spatial ReasoningImage Captioning

LUT: Latent Utility Training for Visual Reasoning

2026-08-01 · Jiaxuan Kang, Siyu Chen, Mingda Li, Mingjie Liu 외 arxiv

Multimodal large language models have advanced visual understanding, yet perception-intensive reasoning remains challenging. Recent latent visual reasoning methods introduce hidden-space computation before answering, but…

Reinforcement LearningVisual Reasoning

Saliency-R1: Enforcing Interpretable and Faithful Vision-language Reasoning via Saliency-map Alignment Reward

2026-04-06 · Shizhan Gong, Minda Hu, Qiyuan Zhang, Chen Ma 외 arxiv

Vision-language models (VLMs) have achieved remarkable success across diverse tasks. However, concerns about their trustworthiness persist, particularly regarding tendencies to lean more on textual cues than visual evide…