玄武：将通用多模态模型演进为内容生态系统工业级基础模型
Xuanwu: Evolving General Multimodal Models into an Industrial-Grade Foundation for Content Ecosystems
AI总结：
本文提出Xuanwu VL-2B，通过三阶段训练流程平衡业务专精与通用能力，实现7项多模态指标平均67.90分，7项业务审核任务召回率94.38%，在对抗性OCR场景中超越Gemini-2.5-Pro。
摘要：
近年来，多模态大模型在通用基准上持续改进。然而，在现实内容审核和对抗场景中，主流模型仍因视觉感知精细度不足和长尾噪声建模不足而出现泛化退化和灾难性遗忘。本文提出Xuanwu VL-2B作为通用多模态模型演进为内容生态系统工业级基础模型的案例。模型采用紧凑的InternViT-300M + MLP + Qwen3 1.7B架构，在约2B参数预算内平衡细粒度视觉感知、语言语义对齐和部署成本。为平衡业务专精与通用能力保留，我们开发了数据迭代和 curated 机制，并通过预训练、中训练和后训练三阶段流程训练模型。消融研究和离线业务评估显示，Xuanwu VL-2B在七项OpenCompass多模态指标上平均得分67.90（对比InternVL 3.5 2B的64.27），在七项独立业务审核任务上平均召回率94.38%，在挑战性对抗性OCR场景中政策违规文本加权整体召回率82.82%，超越Gemini-2.5-Pro（76.72%）。这些结果表明，在有限参数预算下，Xuanwu VL-2B在业务对齐、视觉感知、通用能力保留和部署成本之间实现了实用平衡。
英文摘要：
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.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
