Retrieve, Rerank, Read, then Iterate: Answering Open-Domain Questions of Arbitrary Complexity from Text
By Peng Qi, Haejun Lee, OghenetegiriTGSido, Christopher D. Manning. 2020
17 citations recorded by the source corpus.
Subfields: Retrieval-Augmented Generation, Natural Language Processing
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Retrieval-Augmented Generation
- Keshav Santhanam
- Matei A. Zaharia
- Omar Khattab
- Christopher Potts
- Percy Liang
- Yelong Shen
- Jianfeng Gao
- Xiaodong Liu
In Natural Language Processing
- Jianfeng Gao
- Weizhu Chen
- Xiaodong Liu
- Pengcheng He
- Mohit Bansal
- Percy Liang
- Caiming Xiong
- Luke S. Zettlemoyer
Citations in the corpus
- Adaptive Information Seeking for Open-Domain Question Answering (2021, methodology): “The maximum number of passages inputted into the encoders is set to 3 and the length of input tokens is limited to (Chen et al., 2017) 79.80 - IRRR †∗ (Qi et al., 2020) 84.10 ≥ 150 Asai et al., 2020) 75.70 ≥ 500 HopRetriever †∗ (Li et al., 2021) 512.”
- Reader-Guided Passage Reranking for Open-Domain Question Answering (2021, background): “Another line of work (Das et al., 2018; Qi et al., 2020) reranks the passages by updating the query and often involves a complicated learning process such as R2 interactions.”
- DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines (2023, result): “7 And we test the following custom program, which is akin to Baleen (Khattab et al., 2021a), IRRR (Qi et al., 2020), and has similarities to IRCoT (Trivedi et al., 2022) We can informally compare these with published results for GSM8K.”
- Baleen: Robust Multi-Hop Reasoning at Scale via Condensed Retrieval (2021, background): “For instance, many state-of-the-art systems rely on bag-of-words (Qi et al., 2020) or single-vector dot-product (Xiong et al.”
- Answer Complex Questions: Path Ranker Is All You Need (2021, background): “The first group of methods, such as DDRQA [30] and IRRR [22], conducts retrieval and ranking alternately.”
- Demonstrate-Search-Predict: Composing retrieval and language models for knowledge-intensive NLP (2022, methodology): “This transformation simulates the iterative retrieval component of fine-tuned retrieval-augmented systems like IRRR (Qi et al., 2020), which reads a retrieved passage in every hop and generates a search query (or a termination condition to stop hopping), and Baleen (Khattab et al., 2021a), which…”