Contextualized PACRR for Complex Answer Retrieval

By Sean MacAvaney, Andrew Yates, Kai Hui. 2017

9 citations recorded by the source corpus.

Subfields: Information Retrieval & Search, Question Answering & Reading Comprehension

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Information Retrieval & Search

1. Jimmy J. Lin
2. Rodrigo Nogueira
3. Wei Yang
4. Kyunghyun Cho
5. Zhuyun Dai
6. Zeynep Akkalyoncu Yilmaz
7. Jamie Callan
8. Sebastian Hofstätter

In Question Answering & Reading Comprehension

1. Rodrigo Nogueira
2. Jimmy J. Lin
3. Kyunghyun Cho
4. Jannis Bulian
5. Massimiliano Ciaramita

Citations in the corpus

• Passage Re-ranking with BERT (2019, neutral): “ce. The official pre-trained BERT 1 https://cloud.google.com/tpu/ 2 MS MARCO TREC-CAR MRR@10 MAP Method Dev Eval Test BM25 (Lucene, no tuning) 16.7 16.5 12.3 BM25 (Anserini, tuned) - - 15.3 Co-PACRR? (MacAvaney et al., 2017) - - 14.8 KNRM (Xiong et al., 2017) 21.8 19.8 - Conv-KNRM (Dai et al., 2018) ”
• Multi-Stage Document Ranking with BERT (2019, methodology): “It is also 0.5 MAP points higher than the best entry from TREC 2017 CAR (MacAvaney et al., 2017).”
• Document Expansion by Query Prediction (2019, neutral): “3 650? Co-PACRR♠ (MacAvaney et al., 2017) 14.”
• Multi-agent query reformulation: Challenges and the role of diversity (2019, methodology): “Compared to the top-performing system in the TREC-CAR 2017 Track (MacAvaney et al., 2017), an RL-10-Full with an ensemble of 10 aggregators yields a relative performance improvement of approximately 20%.”