3DrawAgent：通过早期对比经验教 LLM 在 3D 中绘图
3DrawAgent: Teaching LLM to Draw in 3D with Early Contrastive Experience
提出一种无需参数更新的相对经验优化策略，利用 CLIP 感知奖励和 LLM 定性评估构建成对比较，迭代优化 LLM 的 3D 绘图先验，可复用于其他需要黑盒优化的视觉生成任务。
原文依据 a training-free, language-driven framework for 3D sketch generation that leverages large language models (LLMs) to sequentially draw 3D Bezier curves under geometric feedback
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论在三维空间中绘制草图能够对形状、结构和空间关系进行富有表现力的推理，然而通过自然语言生成三维草图仍然是一项重大挑战。本文提出 3DrawAgent，一种免训练、语言驱动的三维草图生成框架，利用大语言模型（LLM）在几何反馈下顺序绘制三维贝塞尔曲线。与先前的二维草图智能体不同，我们的方法引入了一种相对经验优化策略，该策略改编了最近提出的组奖励策略优化（GRPO）范式。该方法不依赖显式的真实标注监督，而是构建生成草图之间的成对比较，每一对包含一个相对更好和一个相对更差的结果，比较基于 CLIP 感知奖励和基于 LLM 的细粒度定性评估。这些经验随后被用于迭代优化三维绘图的先验知识，从而实现模型三维感知能力的黑盒强化。这一设计使我们的模型无需参数更新即可自我提升空间理解和绘图质量。实验表明，3DrawAgent 能够根据多样化的文本提示生成复杂且连贯的三维贝塞尔草图，展现出涌现的几何推理能力，并泛化到新颖形状，为推进免训练三维草图智能领域建立了新范式。
英文原文摘要 1399 字符
Sketching in 3D space enables expressive reasoning about shape, structure, and spatial relationships, yet generating 3D sketches through natural language remains a major challenge. In this work, we introduce 3DrawAgent, a training-free, language-driven framework for 3D sketch generation that leverages large language models (LLMs) to sequentially draw 3D Bezier curves under geometric feedback. Unlike prior 2D sketch agents, our method introduces a relative experience optimization strategy that adapts the recently proposed Group Reward Policy Optimization (GRPO) paradigm. Instead of relying on explicit ground-truth supervision, we construct pairwise comparisons among generated sketches, with each pair consisting of a relatively better and a worse result based on CLIP-based perceptual rewards and LLM-based fine-grained qualitative assessment. These experiences are then used to iteratively refine the prior knowledge of 3D drawing, enabling black-box reinforcement of the model's 3D awareness. This design allows our model to self-improve its spatial understanding and drawing quality without parameter updates. Experiments show that 3DrawAgent can generate complex and coherent 3D Bezier sketches from diverse textual prompts, exhibit emergent geometric reasoning, and generalize to novel shapes, establishing a new paradigm for advancing the field of training-free 3D sketch intelligence.