Generator-retriever-generator: A novel approach to open-domain question answering
A Abdallah, A Jatowt - arXiv preprint arXiv:2307.11278, 2023 - arxiv.org
Open-domain question answering (QA) tasks usually require the retrieval of relevant
information from a large corpus to generate accurate answers. We propose a novel …
information from a large corpus to generate accurate answers. We propose a novel …
Sprint: A unified toolkit for evaluating and demystifying zero-shot neural sparse retrieval
Traditionally, sparse retrieval systems relied on lexical representations to retrieve
documents, such as BM25, dominated information retrieval tasks. With the onset of pre …
documents, such as BM25, dominated information retrieval tasks. With the onset of pre …
Generative retrieval as multi-vector dense retrieval
For a given query generative retrieval generates identifiers of relevant documents in an end-
to-end manner using a sequence-to-sequence architecture. The relation between …
to-end manner using a sequence-to-sequence architecture. The relation between …
Distillation for Multilingual Information Retrieval
Recent work in cross-language information retrieval (CLIR), where queries and documents
are in different languages, has shown the benefit of the Translate-Distill framework that …
are in different languages, has shown the benefit of the Translate-Distill framework that …
Resources for Brewing BEIR: Reproducible Reference Models and Statistical Analyses
BEIR is a benchmark dataset originally designed for zero-shot evaluation of retrieval models
across 18 different domain/task combinations. In recent years, we have witnessed the …
across 18 different domain/task combinations. In recent years, we have witnessed the …
Resources for brewing BEIR: reproducible reference models and an official leaderboard
BEIR is a benchmark dataset for zero-shot evaluation of information retrieval models across
18 different domain/task combinations. In recent years, we have witnessed the growing …
18 different domain/task combinations. In recent years, we have witnessed the growing …
Balanced Knowledge Distillation with Contrastive Learning for Document Re-ranking
Knowledge distillation is commonly used in training a neural document ranking model by
employing a teacher to guide model refinement. As a teacher may not be correct in all cases …
employing a teacher to guide model refinement. As a teacher may not be correct in all cases …
Splate: Sparse late interaction retrieval
The late interaction paradigm introduced with ColBERT stands out in the neural Information
Retrieval space, offering a compelling effectiveness-efficiency trade-off across many …
Retrieval space, offering a compelling effectiveness-efficiency trade-off across many …
End-to-End Retrieval with Learned Dense and Sparse Representations Using Lucene
The bi-encoder architecture provides a framework for understanding machine-learned
retrieval models based on dense and sparse vector representations. Although these …
retrieval models based on dense and sparse vector representations. Although these …
Weighted KL-Divergence for Document Ranking Model Refinement
Transformer-based retrieval and reranking models for text document search are often
refined through knowledge distillation together with contrastive learning. A tight distribution …
refined through knowledge distillation together with contrastive learning. A tight distribution …