1. NCCU Academic Hub
  2. 學術產出
  3. 資訊學院
  4. 期刊論文
  5. Item

學術產出-期刊論文

文章檢視/開啟

書目匯出

Google ScholarTM

政大圖書館

引文資訊

TAIR相關學術產出

題名 Multi-stage Conversational Passage Retrieval: An Approach to Fusing Term Importance Estimation and Neural Query Rewriting
作者 蔡銘峰
Tsai, Ming-Feng
Lin, Sheng-Chieh;Yang, Jheng-Hong;Nogueira, Rodrigo;Wang, Chuan-Ju;Lin, Jimmy
貢獻者 資科系
關鍵詞 Query reformulation; Task models; Environment-specific retrieval
日期 2021-10
上傳時間 2022-10-07
摘要 Conversational search plays a vital role in conversational information seeking. As queries in information seeking dialogues are ambiguous for traditional ad hoc information retrieval (IR) systems due to the coreference and omission resolution problems inherent in natural language dialogue, resolving these ambiguities is crucial. In this article, we tackle conversational passage retrieval, an important component of conversational search, by addressing query ambiguities with query reformulation integrated into a multi-stage ad hoc IR system. Specifically, we propose two conversational query reformulation (CQR) methods: (1) term importance estimation and (2) neural query rewriting. For the former, we expand conversational queries using important terms extracted from the conversational context with frequency-based signals. For the latter, we reformulate conversational queries into natural, stand-alone, human-understandable queries with a pretrained sequence-to-sequence model. Detailed analyses of the two CQR methods are provided quantitatively and qualitatively, explaining their advantages, disadvantages, and distinct behaviors. Moreover, to leverage the strengths of both CQR methods, we propose combining their output with reciprocal rank fusion, yielding state-of-the-art retrieval effectiveness, 30% improvement in terms of NDCG@3 compared to the best submission of Text REtrieval Conference (TREC) Conversational Assistant Track (CAsT) 2019.
關聯 ACM Transactions on Information Systems, 39(4), Article No. 48, pp. 1-29
資料類型 article
DOI https://doi.org/10.1145/3446426
dc.contributor 資科系
dc.creator (作者) 蔡銘峰
dc.creator (作者) Tsai, Ming-Feng
dc.creator (作者) Lin, Sheng-Chieh;Yang, Jheng-Hong;Nogueira, Rodrigo;Wang, Chuan-Ju;Lin, Jimmy
dc.date (日期) 2021-10
dc.date.accessioned 2022-10-07-
dc.date.available 2022-10-07-
dc.date.issued (上傳時間) 2022-10-07-
dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/142230-
dc.description.abstract (摘要) Conversational search plays a vital role in conversational information seeking. As queries in information seeking dialogues are ambiguous for traditional ad hoc information retrieval (IR) systems due to the coreference and omission resolution problems inherent in natural language dialogue, resolving these ambiguities is crucial. In this article, we tackle conversational passage retrieval, an important component of conversational search, by addressing query ambiguities with query reformulation integrated into a multi-stage ad hoc IR system. Specifically, we propose two conversational query reformulation (CQR) methods: (1) term importance estimation and (2) neural query rewriting. For the former, we expand conversational queries using important terms extracted from the conversational context with frequency-based signals. For the latter, we reformulate conversational queries into natural, stand-alone, human-understandable queries with a pretrained sequence-to-sequence model. Detailed analyses of the two CQR methods are provided quantitatively and qualitatively, explaining their advantages, disadvantages, and distinct behaviors. Moreover, to leverage the strengths of both CQR methods, we propose combining their output with reciprocal rank fusion, yielding state-of-the-art retrieval effectiveness, 30% improvement in terms of NDCG@3 compared to the best submission of Text REtrieval Conference (TREC) Conversational Assistant Track (CAsT) 2019.
dc.format.extent 95 bytes-
dc.format.mimetype text/html-
dc.relation (關聯) ACM Transactions on Information Systems, 39(4), Article No. 48, pp. 1-29
dc.subject (關鍵詞) Query reformulation; Task models; Environment-specific retrieval
dc.title (題名) Multi-stage Conversational Passage Retrieval: An Approach to Fusing Term Importance Estimation and Neural Query Rewriting
dc.type (資料類型) article
dc.identifier.doi (DOI) 10.1145/3446426
dc.doi.uri (DOI) https://doi.org/10.1145/3446426