Listwise Generative Retrieval Models via a Sequential Learning Process
文献类型：期刊论文
| 作者 | Tang, Yubao1,2; Zhang, Ruqing1,2; Guo, Jiafeng1,2; De Rijke, Maarten3; Chen, Wei1,2; Cheng, Xueqi1,2 |
| 刊名 | ACM TRANSACTIONS ON INFORMATION SYSTEMS
 |
| 出版日期 | 2024-09-01 |
| 卷号 | 42期号:5页码:31 |
| 关键词 | Document retrieval generative retrieval listwise approach |
| ISSN号 | 1046-8188 |
| DOI | 10.1145/3653712 |
| 英文摘要 | Recently, a novel generative retrieval (GR) paradigm has been proposed, where a single sequence-to-sequence model is learned to directly generate a list of relevant document identifiers (docids) given a query. Existing GR models commonly employ maximum likelihood estimation (MLE) for optimization: This involves maximizing the likelihood of a single relevant docid given an input query, with the assumption that the likelihood for each docid is independent of the other docids in the list. We refer to these models as the pointwise approach in this article. While the pointwise approach has been shown to be effective in the context of GR, it is considered sub-optimal due to its disregard for the fundamental principle that ranking involves making predictions about lists. In this article, we address this limitation by introducing an alternative listwise approach, which empowers the GR model to optimize the relevance at the docid list level. Specifically, we view the generation of a ranked docid list as a sequence learning process: At each step, we learn a subset of parameters that maximizes the corresponding generation likelihood of the ith docid given the (preceding) top i - 1 docids. To formalize the sequence learning process, we design a positional conditional probability for GR. To alleviate the potential impact of beam search on the generation quality during inference, we perform relevance calibration on the generation likelihood of model-generated docids according to relevance grades. We conduct extensive experiments on representative binary and multi-graded relevance datasets. Our empirical results demonstrate that our method outperforms state-of-the-art GR baselines in terms of retrieval performance. |
| 资助项目 | Strategic Priority Research Program of the CAS[XDB0680102] ; National Key Research and Development Program of China[2023YFA1011602] ; National Key Research and Development Program of China[JCKY2022130C039] ; Lenovo-CAS Joint Lab Youth Scientist Project ; CAS Project for Young Scientists in Basic Research[YSBR-034] ; Innovation Project of ICT CAS[E261090] ; Dutch Ministry of Education, Culture and Science through the Netherlands Organisation for Scientific Research - Dutch Research Council (NWO)[NWA.1389.20.183] ; Dutch Ministry of Education, Culture and Science through the Netherlands Organisation for Scientific Research - Dutch Research Council (NWO)[NWA ORC 2020/21] ; European Union's Horizon Europe research and innovation program[101070212] |
| WOS研究方向 | Computer Science |
| 语种 | 英语 |
| WOS记录号 | WOS:001253867000021 |
| 出版者 | ASSOC COMPUTING MACHINERY |
| 源URL | [http://119.78.100.204/handle/2XEOYT63/39846]  |
| 专题 | 中国科学院计算技术研究所期刊论文_英文 |
| 通讯作者 | Guo, Jiafeng |
| 作者单位 | 1.Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China 2.Univ Chinese Acad Sci, 6 Kexueyuan South Rd, Beijing, Peoples R China 3.Univ Amsterdam, Amsterdam, Netherlands |
| 推荐引用方式 GB/T 7714 | Tang, Yubao,Zhang, Ruqing,Guo, Jiafeng,et al. Listwise Generative Retrieval Models via a Sequential Learning Process[J]. ACM TRANSACTIONS ON INFORMATION SYSTEMS,2024,42(5):31. |
| APA | Tang, Yubao,Zhang, Ruqing,Guo, Jiafeng,De Rijke, Maarten,Chen, Wei,&Cheng, Xueqi.(2024).Listwise Generative Retrieval Models via a Sequential Learning Process.ACM TRANSACTIONS ON INFORMATION SYSTEMS,42(5),31. |
| MLA | Tang, Yubao,et al."Listwise Generative Retrieval Models via a Sequential Learning Process".ACM TRANSACTIONS ON INFORMATION SYSTEMS 42.5(2024):31. |
入库方式： OAI收割
来源：计算技术研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。