Establishing Strong Baselines For TripClick Health Retrieval

ADVANCES IN INFORMATION RETRIEVAL, PT II(2022)

引用 5|浏览11
暂无评分
摘要
We present strong Transformer-based re-ranking and dense retrieval baselines for the recently released TripClick health ad-hoc retrieval collection. We improve the - originally too noisy - training data with a simple negative sampling policy. We achieve large gains over BM25 in the re-ranking task of TripClick, which were not achieved with the original baselines. Furthermore, we study the impact of different domain-specific pre-trained models on TripClick. Finally, we show that dense retrieval outperforms BM25 by considerable margins, even with simple training procedures.
更多
查看译文
关键词
retrieval,health,strong baselines
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要