# Structural Evaluation Metrics for SVG Generation via Leave-One-Out Analysis

Source: [Structural Evaluation Metrics for SVG Generation via Leave-One-Out Analysis](<https://www.sota2.com/research/paper/69dc78967d0172140ec78b7e>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-21T05:00:12.672Z

Authors: Haonan Zhu, Adrienne Deganutti, Elad Hirsch, Purvanshi Mehta

Published: 2026-04-09T22:50:41.000Z

## Abstract

SVG generation is typically evaluated by comparing rendered outputs to reference images, which captures visual similarity but not the structural properties that make SVG editable, decomposable, and reusable. Inspired by the classical jackknife, we introduce element-level leave-one-out (LOO) analysis. The procedure renders the SVG with and without each element, which yields element-level signals for quality assessment and structural analysis. From this single mechanism, we derive (i) per-element quality scores that enable zero-shot artifact detection; (ii) element-concept attribution via LOO footprints crossed with VLM-grounded concept heatmaps; and (iii) four structural metrics: purity, coverage, compactness, and locality, which quantify SVG modularity from complementary angles. These metrics extend SVG evaluation from image similarity to code structure, enabling element-level diagnosis and comparison of how visual concepts are represented, partitioned, and organized within SVG code. Their practical relevance is validated on over 19,000 edits (5 types) across 5 generation systems and 3 complexity tiers.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [SVG Edit Precision on SVG Complexity Dataset complex tier](<https://www.sota2.com/research/sota/svg-edit-precision-on-svg-complexity-dataset-complex-tier.md>): SVG Edit Precision; SVG Complexity Dataset complex tier; Overall Precision: not recorded.

- [SVG Generation on SVG complex tier](<https://www.sota2.com/research/sota/svg-generation-on-svg-complex-tier.md>): SVG Generation; SVG complex tier; Purity: not recorded.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2604.08809>)

[PDF](<https://arxiv.org/pdf/2604.08809.pdf>)
