SVG-BERT, a code-native encoder pretrained directly on SVG source code as representation infrastructure for structure-aware vector design workflows, consistently outperforms generic text encoders and provides information complementary to raster encoders, which remain stronger on strongly visual semantic tasks.
VectorHarness is presented, a multi-agent framework for raster-to-authoring reconstruction that recovers heterogeneous components using type-appropriate native representations that recovers heterogeneous components using type-appropriate native representations.
Jia-Hao Tang, Yi-Ren Song, Alex Jinpeng Wang· 0 citations
This work proposes a VLM-driven agentic framework for semantic compositional SVG generation that recursively parses visual scenes into semantic and geometric hierarchies via top-down decomposition, visual grounding, and prompt-driven amodal occlusion recovery, ensuring each component is geometrically complete.
Sehwan Park, Taehoon Kim, Geonhee Han et al.· 0 citations
SVGEval is introduced, a vision-grounded multimodal benchmark for human-aligned SVG quality assessment that incorporates visual renderings to evaluate whether models can judge the rendered outcome rather than only inspect SVG code, and provides high-quality annotations obtained via multi-round human labeling with exper...
Yimin Wang, Ye Chen, Hanqi Chen et al.· 1 citation
This work introduces text-guided vector object extraction, where the goal is to recover an editable sub-SVG corresponding to a natural-language query, and proposes a raster-grounded extraction method that localizes the queried object in the rendered image and maps the resulting region back to SVG primitives through ove...
Ivan Jarsky, Boris Malashenko, B. Timofeenko et al.· 0 citations
This work introduces VisEditBench, a benchmark of 1,395 human-annotated visualization code-editing tasks grounded in realistic visualization workflows and failure cases, and proposes VisEditAgent, a render-grounded editing framework that iteratively generates, executes, validates, and refines candidate edits.
This work introduces a human-aligned evaluation framework for text-to-SVG generation, and develops two complementary evaluators: CLIP scorers adapted to vector graphics and then aligned to human preferences, for fast large-scale evaluation and a VLM judge trained with supervised fine-tuning and reward-shaped reinforcem...
Marco Cipriano, Leonardo Zini, Alexandra Schild et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
