VARCO-VISION-2.0 Technical Report

📅 2025-09-12
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
Existing bilingual vision-language models (VLMs) exhibit limitations in multi-image understanding and layout-aware OCR—i.e., joint modeling of textual content and its spatial coordinates. To address this, we propose VARCO-VISION-2.0, the first open-source Korean–English bilingual VLM. Methodologically, it introduces a novel four-stage curriculum learning strategy coupled with a memory-efficient training framework, integrating multimodal alignment enhancement, language capability preservation, safety optimization, and spatially aware OCR. Key contributions include: (1) robust bilingual understanding of complex multi-image inputs—including documents, charts, and tables; (2) release of two model variants—14B and 1.7B parameters—balancing high performance with edge-device deployability; and (3) strong empirical performance: the 14B model ranks eighth among same-scale models on the OpenCompass VLM leaderboard, with significant gains in bilingual spatial grounding accuracy.

Technology Category

Application Category

📝 Abstract
We introduce VARCO-VISION-2.0, an open-weight bilingual vision-language model (VLM) for Korean and English with improved capabilities compared to the previous model VARCO-VISION-14B. The model supports multi-image understanding for complex inputs such as documents, charts, and tables, and delivers layoutaware OCR by predicting both textual content and its spatial location. Trained with a four-stage curriculum with memory-efficient techniques, the model achieves enhanced multimodal alignment, while preserving core language abilities and improving safety via preference optimization. Extensive benchmark evaluations demonstrate strong spatial grounding and competitive results for both languages, with the 14B model achieving 8th place on the OpenCompass VLM leaderboard among models of comparable scale. Alongside the 14B-scale model, we release a 1.7B version optimized for on-device deployment. We believe these models advance the development of bilingual VLMs and their practical applications. Two variants of VARCO-VISION-2.0 are available at Hugging Face: a full-scale 14B model and a lightweight 1.7B model.
Problem

Research questions and friction points this paper is trying to address.

Develops bilingual vision-language model for Korean and English
Enables multi-image understanding of documents and charts
Delivers layout-aware OCR with spatial content prediction
Innovation

Methods, ideas, or system contributions that make the work stand out.

Open-weight bilingual vision-language model
Multi-image understanding with layout-aware OCR
Four-stage curriculum training with preference optimization
🔎 Similar Papers
No similar papers found.
Y
Young-rok Cha
NC AI
J
Jeongho Ju
NC AI
S
SunYoung Park
NC AI
J
Jong-Hyeon Lee
NC AI
Y
Younghyun Yu
NC AI
Y
Youngjune Kim
NC AI