---
格式版本: 2
标题: "BPG: Balancing Plasticity and Generalization for Domain Incremental Learning"
原文链接: "https://arxiv.org/abs/2608.10804"
发布日期: "2026-08-11"
发布时间校准状态: "found"
发布时间需复核: "否"
发布时间来源: "rule:local:strict_original_body"
发布时间证据: "**\\[v1\\]** Tue, 11 Aug 2026 11:22:27 UTC (2,996 KB)"
发布时间校准原因: "规则确认唯一严格发布时间，来源 local:strict_original_body"
发布时间校准置信度: "high"
发布时间候选数量: 18
发布时间严格候选数量: 6
发布时间原页读取状态: "source template page reused from URL open"
发布时间未找到原因: ""
发布时间校准时间: "2026-08-12T11:27:27+08:00"
发布时间仲裁状态: "skipped"
发布时间仲裁尝试次数: 0
发布时间仲裁耗时毫秒: 0
发现时间: "2026-08-12T11:25:10+08:00"
入库时间: "2026-08-12T03:27:27.934Z"
来源平台: "arXiv 学术论文搜索"
搜索渠道: "source_template"
搜索词: "https://arxiv.org/search/?query=model%20fit&searchtype=all"
匹配关键词:
  - "model fit"
  - "performance"
  - "AI"
相关厂家:
  []
相关专家:
  []
内容类型: "网页"
抓取工具: "Free Fetch + Defuddle"
清洗工具: "Defuddle Markdown + Defuddle/Readability 正文提取"
原始附件:
  []
AI优质: "否"
AI打分: 0
AI分档: "非优质"
AI质检状态: "不通过"
AI打分理由: "主题与超节点/AI Rack/机柜级AI基础设施等完全无关，属于计算机视觉领域论文，不符合项目关注范围。"
AI质检模型: "ali-deepseek-v4-flash"
AI质检时间: "2026-08-12T11:27:44+08:00"
AI主题相关性: 0
AI来源权威性: 0
AI新颖性: 0
AI技术细节: 0
AI商业部署信号: 0
AI完整性: 0
采集批次: "2026年8月12日2点04分20秒"
采集批次ID: "20260812-020420-289"
去重键: "https://arxiv.org/abs/2608.10804"
---

## Computer Science > Computer Vision and Pattern Recognition

## Title:BPG: Balancing Plasticity and Generalization for Domain Incremental Learning

Authors:[Qiang Wang](https://arxiv.org/search/cs?searchtype=author&query=Wang,+Q), [Songlin Dong](https://arxiv.org/search/cs?searchtype=author&query=Dong,+S), [Shaokun Wang](https://arxiv.org/search/cs?searchtype=author&query=Wang,+S), [Jizhou Han](https://arxiv.org/search/cs?searchtype=author&query=Han,+J), [Xiang Song](https://arxiv.org/search/cs?searchtype=author&query=Song,+X), [Chenhao Ding](https://arxiv.org/search/cs?searchtype=author&query=Ding,+C), [Yuhang He](https://arxiv.org/search/cs?searchtype=author&query=He,+Y), [Yihong Gong](https://arxiv.org/search/cs?searchtype=author&query=Gong,+Y)

[View PDF](https://arxiv.org/pdf/2608.10804) [HTML (experimental)](https://arxiv.org/html/2608.10804v1)

> Abstract:Deep neural networks excel in various tasks but struggle to generalize across evolving data distributions, leading to significant performance degradation under domain shifts. Domain incremental learning (DIL) addresses this challenge by enabling models to continuously adapt while retaining prior knowledge. Among existing DIL approaches, the parameter-isolation paradigm achieves state-of-the-art performance. However, these methods often adopt a one-size-fits-all approach to adapt to new domains, resulting in either insufficient learning capacity or redundant parameters. In this work, we propose BPG, a unified framework that addresses both challenges through two complementary components: BPG-Adapter, which dynamically determines each domain's adapter hidden dimension based on domain-specific feature separability, and BPG-Inference, a soft domain mixture strategy that integrates multiple domain-specific models at test time, mitigating domain ID misselection. Experimental results on DomainNet, CDDB, and CORe50 demonstrate that BPG consistently outperforms uniform adapter-based approaches and hard domain selection strategies, achieving state-of-the-art average accuracy while reducing forgetting to as low as 0.22% on DomainNet.

| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| --- | --- |
| Cite as: | [arXiv:2608.10804](https://arxiv.org/abs/2608.10804) \[cs.CV\] |
|  | (or [arXiv:2608.10804v1](https://arxiv.org/abs/2608.10804v1) \[cs.CV\] for this version) |
|  | [https://doi.org/10.48550/arXiv.2608.10804](https://doi.org/10.48550/arXiv.2608.10804) |

## Submission history

From: Qiang Wang \[[view email](https://arxiv.org/show-email/c754ccdb/2608.10804)\]  
**\[v1\]** Tue, 11 Aug 2026 11:22:27 UTC (2,996 KB)

[Which authors of this paper are endorsers?](https://arxiv.org/auth/show-endorsers/2608.10804) | Disable MathJax ([What is MathJax?](https://info.arxiv.org/help/mathjax.html))