Chart2Code-MoLA: Efficient Multi-Modal Code Generation via Adaptive Expert Routing

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Yifei, Keung, Jacky, Mao, Zhenyu, Zhang, Jingyu, Cao, Yuchen
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918222666661888
author Wang, Yifei
Keung, Jacky
Mao, Zhenyu
Zhang, Jingyu
Cao, Yuchen
author_facet Wang, Yifei
Keung, Jacky
Mao, Zhenyu
Zhang, Jingyu
Cao, Yuchen
contents Chart-to-code generation is a critical task in automated data visualization, translating complex chart structures into executable programs. While recent Multi-modal Large Language Models (MLLMs) improve chart representation, existing approaches still struggle to achieve cross-type generalization, memory efficiency, and modular design. To address these challenges, this paper proposes C2C-MoLA, a multimodal framework that synergizes Mixture of Experts (MoE) with Low-Rank Adaptation (LoRA). The MoE component uses a complexity-aware routing mechanism with domain-specialized experts and load-balanced sparse gating, dynamically allocating inputs based on learnable structural metrics like element count and chart complexity. LoRA enables parameter-efficient updates for resource-conscious tuning, further supported by a tailored training strategy that aligns routing stability with semantic accuracy. Experiments on Chart2Code-160k show that the proposed model improves generation accuracy by up to 17%, reduces peak GPU memory by 18%, and accelerates convergence by 20%, when compared to standard fine-tuning and LoRA-only baselines, particularly on complex charts. Ablation studies validate optimal designs, such as 8 experts and rank-8 LoRA, and confirm scalability for real-world multimodal code generation.
format Preprint
id arxiv_https___arxiv_org_abs_2511_23321
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Chart2Code-MoLA: Efficient Multi-Modal Code Generation via Adaptive Expert Routing
Wang, Yifei
Keung, Jacky
Mao, Zhenyu
Zhang, Jingyu
Cao, Yuchen
Software Engineering
Chart-to-code generation is a critical task in automated data visualization, translating complex chart structures into executable programs. While recent Multi-modal Large Language Models (MLLMs) improve chart representation, existing approaches still struggle to achieve cross-type generalization, memory efficiency, and modular design. To address these challenges, this paper proposes C2C-MoLA, a multimodal framework that synergizes Mixture of Experts (MoE) with Low-Rank Adaptation (LoRA). The MoE component uses a complexity-aware routing mechanism with domain-specialized experts and load-balanced sparse gating, dynamically allocating inputs based on learnable structural metrics like element count and chart complexity. LoRA enables parameter-efficient updates for resource-conscious tuning, further supported by a tailored training strategy that aligns routing stability with semantic accuracy. Experiments on Chart2Code-160k show that the proposed model improves generation accuracy by up to 17%, reduces peak GPU memory by 18%, and accelerates convergence by 20%, when compared to standard fine-tuning and LoRA-only baselines, particularly on complex charts. Ablation studies validate optimal designs, such as 8 experts and rank-8 LoRA, and confirm scalability for real-world multimodal code generation.
title Chart2Code-MoLA: Efficient Multi-Modal Code Generation via Adaptive Expert Routing
topic Software Engineering
url https://arxiv.org/abs/2511.23321