# Reliable Reasoning in SVG-LLMs via Multi-Task Multi-Reward Reinforcement Learning

arXiv: https://arxiv.org/abs/2603.16189 · HTML: https://mlsanity.com/paper/2603.16189/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Placed in Multimodal reasoning, one of 17 directions in LLM reasoning.
- Builds on 10 of its 45 references that are in the map.
- First posted on arXiv 17 Mar 2026.

## Placed in

- LLM reasoning / [Multimodal reasoning](https://mlsanity.com/lr/multimodal-reasoning/)
- Multimodal LLMs / vision-language models / [Chart understanding and anomaly detection](https://mlsanity.com/mllm/chart-understanding-and-anomaly-detection/)

## Builds on (most cited first, 10 of 10)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [DreamFusion: Text-to-3D using 2D Diffusion](https://mlsanity.com/paper/2209.14988/) (2022)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models](https://mlsanity.com/paper/2504.10479/) (2025)
- [InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency](https://mlsanity.com/paper/2508.18265/) (2025)
- [A Neural Representation of Sketch Drawings](https://mlsanity.com/paper/1704.03477/) (2017)
- [StyleCLIPDraw: Coupling Content and Style in Text-to-Drawing Translation](https://mlsanity.com/paper/2202.12362/) (2022)
- [Reason-SVG: Enhancing Structured Reasoning for Vector Graphics Generation with Reinforcement Learning](https://mlsanity.com/paper/2505.24499/) (2025)

## Built on by (most cited first, 6 of 6)

- [Render-in-the-Loop: Vector Graphics Generation via Visual Self-Feedback](https://mlsanity.com/paper/2604.20730/) (2026)
- [GeoSVG-RL: Geometry-Aware Reinforcement Learning for Layout-Constrained Text-to-SVG Diagram Generation](https://mlsanity.com/paper/2605.25447/) (2026)
- [Structural Evaluation Metrics for SVG Generation via Leave-One-Out Analysis](https://mlsanity.com/paper/2604.08809/) (2026)
- [GVR-Coder: A Visual-Feedback Framework for Structured SVG Generation in Complex Document and Meeting Scenarios](https://mlsanity.com/paper/2607.28073/) (2026)
- [Not All Prompts Are Equal: Exploration-Guided Prompt Scaffolding for Multimodal Reinforcement Post-Training](https://mlsanity.com/paper/2609.15051/) (2026)
- [Reinforcement Learning from Intermediate Renders for Image-to-Code Generation](https://mlsanity.com/paper/2609.34587/) (2026)

## Abstract (arXiv)

With the rapid advancement of vision-language models, an increasing number of studies have explored their potential for SVG generation tasks. Although existing approaches improve performance by constructing large-scale SVG datasets and introducing SVG-specific tokens, they still suffer from limited generalization, redundant paths in code outputs, and a lack of explicit reasoning. In this work, we present CTRL-S (Chain-of-Thought Reinforcement Learning for SVG), a unified framework that introduces a chain-of-thought mechanism to explicitly expose the model's reasoning process during SVG generation. To support this structured reasoning, we construct SVG-Sophia, a high-quality dataset containing 145K samples across SVG code refinement, Text-to-SVG, and Image-to-SVG tasks. By training the model to generate group-level structured SVG code, CTRL-S significantly improves structural coherence and visual fidelity. Furthermore, we adopt the GRPO algorithm and design a multi-reward optimization framework, incorporating DINO, image-text similarity, format, and code efficiency rewards. Through joint multi-reward optimization and multi-task training, our approach systematically enhances overall generation capabilities. Extensive experiments show that CTRL-S outperforms existing methods, achieving higher task success rates, superior SVG code quality, and exceptional visual fidelity.