J
Jiaxin Mao
Researcher at Renmin University of China
Publications - 109
Citations - 1475
Jiaxin Mao is an academic researcher from Renmin University of China. The author has contributed to research in topics: Ranking (information retrieval) & Computer science. The author has an hindex of 15, co-authored 78 publications receiving 578 citations. Previous affiliations of Jiaxin Mao include Tsinghua University.
Papers
More filters
Proceedings ArticleDOI
Optimizing Dense Retrieval Model Training with Hard Negatives
TL;DR: Zhang et al. as mentioned in this paper investigated different training strategies for dense retrieval models and tried to explain why hard negative sampling performs better than random sampling, and proposed two training strategies named a stable training algorithm for dense retrieval (STAR) and a query-side training Algorithm for Directly Optimizing Ranking pErformance (ADORE), respectively.
Posted Content
RepBERT: Contextualized Text Embeddings for First-Stage Retrieval.
TL;DR: This work proposes a different approach, called RepBERT, to represent documents and queries with fixed-length contextualized embeddings, which achieves state-of-the-art results among all initial retrieval techniques.
Proceedings ArticleDOI
When does Relevance Mean Usefulness and User Satisfaction in Web Search
Jiaxin Mao,Yiqun Liu,Ke Zhou,Jian-Yun Nie,Jingtao Song,Min Zhang,Shaoping Ma,Jiashen Sun,Hengliang Luo +8 more
TL;DR: It is shown that external assessors are capable of annotating usefulness when provided with more search context information and it is suggested that a usefulness-based evaluation method can be defined to better reflect the quality of search systems perceived by the users.
Proceedings ArticleDOI
BERT-PLI: Modeling Paragraph-Level Interactions for Legal Case Retrieval
TL;DR: BERT-PLI is proposed, a novel model that utilizes BERT to capture the semantic relationships at the paragraph-level and then infers the relevance between two cases by aggregating paragraph- level interactions.
Posted Content
Neural Logic Reasoning
TL;DR: Logic-Integrated Neural Network (LINN) is a dynamic neural architecture that builds the computational graph according to input logical expressions and learns basic logical operations such as AND, OR, NOT as neural modules, and conducts propositional logical reasoning through the network for inference.