CVJun 30

UniCoder: Unified Visual-to-Code Generation via Symbolic Rewards and Reference-Guided Code Optimization

arXiv:2606.3173215.9Has Code
Predicted impact top 14% in CV · last 90 daysOriginality Highly original
AI Analysis

This work addresses the critical problem of fine-grained alignment in visual-to-code generation for scientific and design applications, where existing MLLMs fail.

UniCoder introduces a unified RL framework with Symbolic Attribute Alignment and Reference-Guided Code Optimization to achieve pixel-precise visual-to-code generation, surpassing all open-source baselines and matching proprietary models on four benchmarks.

Visual-to-Code generation, which transforms scientific plots, vector graphics, and webpages into executable scripts, demands a level of pixel-precise alignment that standard Multimodal Large Language Models (MLLMs) fail to achieve through Supervised Fine-Tuning (SFT) alone. While Reinforcement Learning (RL) offers a theoretical pathway to bridge this gap, its application is hindered by two fundamental obstacles: (1) \textit{Reward Coarseness}, where semantic metrics like CLIP scores fail to penalize fine-grained element deviations, and (2) \textit{Exploration Stagnation}, where the sparse, heterogeneous code search space prevents the policy from bootstrapping valid trajectories. To overcome these limitations, we introduce UniCoder, a unified RL framework that integrates two novel mechanisms. First, we propose \textbf{Symbolic Attribute Alignment}, which employs a lightweight auxiliary LLM to parse generated code into discrete visual attributes (e.g., hex colors, coordinate limits), enabling dense, element-wise reward computation. Second, to escape local optima, we devise \textbf{Reference-Guided Code Optimization}, a strategy that dynamically injects ground-truth trajectories into low-performing rollout groups, transforming blind exploration into guided policy improvement. Extensive experiments on ChartMimic, UniSVG, Design2Code and ScreenBench benchmarks demonstrate that our 8B-parameter model not only surpasses all open-source baselines but also achieves state-of-the-art performance comparable to proprietary models, establishing a new paradigm for generalized visual-to-code synthesis.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes