测试搜索系统随时间推移的性能退化，发现2023年表现最佳。
DS@GT at LongEval: Evaluating Temporal Performance in Web Search Systems and Topics with Two-Stage Retrieval
- 分两阶段检索：先关键词匹配，再扩展查询与重排序。
- 平均NDCG@10达0.296，2023年5月最高达0.395。
- 适合关注长期检索性能评估的研究者。
信息检索模型通常在静态数据集上训练，随网页内容演变易出现性能下降。DS@GT团队参与了CLEF 2025举办的长时序模型性能评估（LongEval）任务，评估检索系统在不同时期网络快照下的表现。基于Qwant网络数据集，我们进行了随时间演化的主题建模分析。所提出的两阶段检索系统采用稀疏关键词搜索，结合查询扩展与文档重排序。最佳系统在全部训练与测试数据上的平均NDCG@10为0.296，2023年5月的最高得分为0.395。相关代码已开源。
原文摘要 · Abstract (English)
Information Retrieval (IR) models are often trained on static datasets, making them vulnerable to performance degradation as web content evolves. The DS@GT competition team participated in the Longitudinal Evaluation of Model Performance (LongEval) lab at CLEF 2025, which evaluates IR systems across temporally distributed web snapshots. Our analysis of the Qwant web dataset includes exploratory data analysis with topic modeling over time. The two-phase retrieval system employs sparse keyword searches, utilizing query expansion and document reranking. Our best system achieves an average NDCG@10 of 0.296 across the entire training and test dataset, with an overall best score of 0.395 on 2023-05. The accompanying source code for this paper is at https://github.com/dsgt-arc/longeval-2025
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。