VisionFoundry: Teaching VLMs Visual Perception with Synthetic Images

2026-04-10 · Guanyu Zhou, Yida Yin, Wenhao Chai, Shengbang Tong, Xingyu Fu, Zhuang Liu arxiv

Vision-language models (VLMs) still struggle with visual perception tasks such as spatial understanding and viewpoint recognition. One plausible contributing factor is that natural image datasets provide limited supervision for low-level visual skills. This motivates a practical question: can targeted synthetic supervision, generated from only a task keyword such as Depth Order, address these weaknesses? To investigate this question, we introduce VisionFoundry, a task-aware synthetic data generation pipeline that takes only the task name as input and uses large language models (LLMs) to generate questions, answers, and text-to-image (T2I) prompts, then synthesizes images with T2I models and verifies consistency with a proprietary VLM, requiring no reference images or human annotation. Using VisionFoundry, we construct VisionFoundry-10K, a synthetic visual question answering (VQA) dataset containing 10k image-question-answer triples spanning 10 tasks. Models trained on VisionFoundry-10K achieve substantial improvements on visual perception benchmarks: +7% on MMVP and +10% on CV-Bench-3D, while preserving broader capabilities and showing favorable scaling behavior as data size increases. Our results suggest that limited task-targeted supervision is an important contributor to this bottleneck and that synthetic supervision is a promising path toward more systematic training for VLMs.

📄 PDF Abstract BibTeX arXiv:2604.09531

Code (0)

등록된 구현이 없습니다.

Tasks

Synthetic Data Generation Visual Question Answering

VLMs Need Words: Vision Language Models Ignore Visual Detail In Favor of Semantic Anchors

2026-04-02 · Haz Sameen Shahgir, Xiaofu Chen, Yu Fu, Erfan Shayegani 외 arxiv

Vision-language models (VLMs) have achieved impressive performance across a wide range of multimodal tasks. However, they often fail on tasks that require fine-grained visual perception, even when the required informatio…

Semantic correspondence Multimodal Reasoning

MVP-Bench: Can Large Vision--Language Models Conduct Multi-level Visual Perception Like Humans?

2024-10-06 · Guanzhen Li, Yuxi Xie, Min-Yen Kan

Humans perform visual perception at multiple levels, including low-level object recognition and high-level semantic interpretation such as behavior understanding. Subtle differences in low-level details can lead to subst…

Object Recognition

Chain-of-Visual-Thought: Teaching VLMs to See and Think Better with Continuous Visual Tokens

2025-11-24 · Yiming Qin, Bomin Wei, Jiaxin Ge, Konstantinos Kallidromitis 외 arxiv

Vision-Language Models (VLMs) excel at reasoning in linguistic space but struggle with perceptual understanding that requires dense visual perception, e.g., spatial reasoning and geometric awareness. This limitation stem…

Spatial Reasoning

GeoMeter: Probing Depth and Height Perception of Large Visual-Language Models

2024-08-21 · Shehreen Azad, Yash Jain, Rishit Garg, Yogesh S Rawat 외

Geometric understanding is crucial for navigating and interacting with our environment. While large Vision Language Models (VLMs) demonstrate impressive capabilities, deploying them in real-world scenarios necessitates a…

Unleashing Perception-Time Scaling to Multimodal Reasoning Models

2025-10-10 · Yifan Li, Zhenghao Chen, Ziheng Wu, Kun Zhou 외 arxiv

Recent advances in inference-time scaling, particularly those leveraging reinforcement learning with verifiable rewards, have substantially enhanced the reasoning capabilities of Large Vision-Language Models (LVLMs). Ins…

Reinforcement Learning Multimodal Reasoning