通过视觉反馈动态调整采样，让文本生成复杂矢量图更连贯。
CANVAS: Consistency-Aware Navigation via Visual Adaptive Sampling for Long-Context Text-to-SVG Generation
- 基于渲染结果反馈，动态分配生成资源，优化每一步笔画选择。
- 在不训练模型的前提下，显著提升图形几何、布局和遮挡关系的一致性。
- 适合需要高一致性矢量图生成的场景，如复杂图标与插画设计。
自回归大模型已将文本到矢量图（Text-to-SVG）生成从简单图标推进到复杂长上下文图形，但标准自回归解码常无法维持几何、布局、遮挡和构图等全局一致性。本文提出CANVAS（Consistency-Aware Navigation via Visual Adaptive Sampling），一种无需训练、感知渲染结果的推理框架，结合幂强化轨迹似然与渲染未来图像的视觉反馈，推导出逐笔画导航规则。该方法在有限生成与渲染预算下，有效评估每条候选笔画的未来价值，并根据候选不确定性、决策影响和回溯成本自适应分配采样。在多个自回归SVG骨干模型及互补基准上实验表明，该框架在不增加训练成本的情况下，显著提升了图形的全局一致性，包括合理的几何关系、空间布局、遮挡顺序与整体构图，验证了其有效性与泛化能力。
原文摘要 · Abstract (English)
Autoregressive large models have recently advanced Text-to-SVG generation from simple icons to complex, long-context graphics, yet standard autoregressive decoding often fails to maintain global consistency across geometry, layout, occlusion, and composition. We introduce CANVAS (Consistency-Aware Navigation via Visual Adaptive Sampling), a training-free, render-aware inference framework that combines power-sharpened trajectory likelihood with visual feedback from rendered futures and derives a stroke-wise navigation rule. It effectively estimates each candidate stroke's future value under a limited generation and rendering budget and adaptively allocates samples according to candidate uncertainty, decision influence, and rollout cost. Experiments across multiple autoregressive SVG backbones and complementary benchmarks demonstrate improvements in global consistency, which includes sound geometric relationships, spatial layouts, occlusion ordering, and overall composition, without additional training, demonstrating the effectiveness and generalization ability of our framework.
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。