Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Xuanwu: Evolving General Multimodal Models into an Industrial-Grade Foundation for Content Ecosystems

Zhiqian Zhang, Xu Zhao, Xiaoqing Xu, Guangdong Liang, Weijia Wang, Xiaolei Lv, Bo Li, Jun Gao

Abstract

In recent years, multimodal large models have continued to improve on general benchmarks. However, in real-world content moderation and adversarial settings, mainstream models still suffer from degraded generalization and catastrophic forgetting because of limited fine-grained visual perception and insufficient modeling of long-tail noise. In this paper, we present Xuanwu VL-2B as a case study of how general multimodal models can be developed into an industrial-grade foundation model for content ecosystems. The model adopts a compact InternViT-300M + MLP + Qwen3 1.7B architecture, balancing fine-grained visual perception, language-semantic alignment, and deployment cost within an approximately 2B-parameter budget. To balance business specialization with the retention of general capabilities, we developed a data iteration and curation mechanism and trained the model through a progressive three-stage pipeline: pre-training, mid-training, and post-training. Ablation studies and offline business evaluations show that Xuanwu VL-2B achieves an average score of 67.90 across seven OpenCompass multimodal metrics (vs. 64.27 for InternVL 3.5 2B), an average recall of 94.38% over seven independent business moderation tasks, and a weighted overall recall of 82.82% on policy-violating text in challenging adversarial OCR scenarios, outperforming Gemini-2.5-Pro (76.72%). These results show that, under a limited parameter budget, Xuanwu VL-2B achieves a practical balance among business alignment, visual perception, general capability retention, and deployment cost.

Xuanwu: Evolving General Multimodal Models into an Industrial-Grade Foundation for Content Ecosystems

Abstract

In recent years, multimodal large models have continued to improve on general benchmarks. However, in real-world content moderation and adversarial settings, mainstream models still suffer from degraded generalization and catastrophic forgetting because of limited fine-grained visual perception and insufficient modeling of long-tail noise. In this paper, we present Xuanwu VL-2B as a case study of how general multimodal models can be developed into an industrial-grade foundation model for content ecosystems. The model adopts a compact InternViT-300M + MLP + Qwen3 1.7B architecture, balancing fine-grained visual perception, language-semantic alignment, and deployment cost within an approximately 2B-parameter budget. To balance business specialization with the retention of general capabilities, we developed a data iteration and curation mechanism and trained the model through a progressive three-stage pipeline: pre-training, mid-training, and post-training. Ablation studies and offline business evaluations show that Xuanwu VL-2B achieves an average score of 67.90 across seven OpenCompass multimodal metrics (vs. 64.27 for InternVL 3.5 2B), an average recall of 94.38% over seven independent business moderation tasks, and a weighted overall recall of 82.82% on policy-violating text in challenging adversarial OCR scenarios, outperforming Gemini-2.5-Pro (76.72%). These results show that, under a limited parameter budget, Xuanwu VL-2B achieves a practical balance among business alignment, visual perception, general capability retention, and deployment cost.

Paper Structure

This paper contains 50 sections, 3 equations, 10 figures, 18 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Model Architecture
  4. Architecture Design
  5. Vision Encoder Exploration and Selection
  6. Language Foundation and Cross-Modal Alignment
  7. Architecture Hyperparameter Details
  8. Dynamic High-Resolution Perception Mechanism
  9. Three-Stage Training
  10. Pre-Training
  11. Mid-Training
  12. Post-Training
  13. Supervised Fine-Tuning
  14. Reinforcement Learning
  15. Explainability and Chain-of-Thought
  16. ...and 35 more sections

Figures (10)

  • Figure 1: Capability layering of the Xuanwu industrial multimodal foundation. The lower layer provides unified multimodal representations, while the upper layers address moderation and adversarial defense.
  • Figure 2: Core architecture of Xuanwu VL-2B: InternViT-300M serves as the visual backbone and Qwen3 1.7B serves as the language backbone, connected by an MLP projector.
  • Figure 3: Handwriting-variant case: cursive diversion codes hidden in a complex product image
  • Figure 4: Layered-watermark case: low-opacity contact information hidden in the image
  • Figure 5: Micro-font case: miniature variant-character diversion
  • ...and 5 more figures