用文本生成难例，自动提升数据多样性。
Towards Continual Expansion of Data Coverage: Automatic Text-guided Edge-case Synthesis
- 用微调的LLM将图像描述转为多样提示，引导生成难样本。
- 在FishEye8K上检测鲁棒性超越人工和普通增强方法。
- 适合需要持续改进数据质量的视觉系统开发者。
深度神经网络性能高度依赖训练数据质量。然而，通过人工筛选挑战性边缘案例来缓解数据偏差仍是主要瓶颈。为此，我们提出一种自动化文本引导的边缘案例合成流程。该方法利用经过偏好学习微调的大语言模型，将图像描述重写为多样化的文本提示，引导文生图模型生成具有挑战性的视觉场景。在FishEye8K目标检测基准上的评估显示，该方法显著提升了模型鲁棒性，优于原始增强和人工设计的提示。本工作建立了一个可扩展框架，将数据整理从手动转向自动化、精准合成，为构建更可靠且持续进化的AI系统提供了新方向。代码已开源：https://github.com/gokyeongryeol/ATES。
原文摘要 · Abstract (English)
The performance of deep neural networks is strongly influenced by the quality of their training data. However, mitigating dataset bias by manually curating challenging edge cases remains a major bottleneck. To address this, we propose an automated pipeline for text-guided edge-case synthesis. Our approach employs a Large Language Model, fine-tuned via preference learning, to rephrase image captions into diverse textual prompts that steer a Text-to-Image model toward generating difficult visual scenarios. Evaluated on the FishEye8K object detection benchmark, our method achieves superior robustness, surpassing both naive augmentation and manually engineered prompts. This work establishes a scalable framework that shifts data curation from manual effort to automated, targeted synthesis, offering a promising direction for developing more reliable and continuously improving AI systems. Code is available at https://github.com/gokyeongryeol/ATES.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。