Unsupervised Alignment-based Iterative Evidence Retrieval for Multi-hop Question Answering

2020-05-04 · ACL 2020 6 · Vikas Yadav, Steven Bethard, Mihai Surdeanu

Evidence retrieval is a critical stage of question answering (QA), necessary not only to improve performance, but also to explain the decisions of the corresponding QA method. We introduce a simple, fast, and unsupervised iterative evidence retrieval method, which relies on three ideas: (a) an unsupervised alignment approach to soft-align questions and answers with justification sentences using only GloVe embeddings, (b) an iterative process that reformulates queries focusing on terms that are not covered by existing justifications, which (c) a stopping criterion that terminates retrieval when the terms in the given question and candidate answers are covered by the retrieved justifications. Despite its simplicity, our approach outperforms all the previous methods (including supervised methods) on the evidence selection task on two datasets: MultiRC and QASC. When these evidence sentences are fed into a RoBERTa answer classification component, we achieve state-of-the-art QA performance on these two datasets.

📄 PDF Abstract BibTeX arXiv:2005.01218

Code (1)

vikas95/AIR-retriever 공식 구현

Tasks

Evidence SelectionMulti-hop Question Answering Question Answering Retrieval

Methods 이 논문이 사용한 방법론

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.

Adam 설명 없음

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…

Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…

Attention 설명 없음

Multi-Head Attention 설명 없음

Residual Connection 설명 없음

CORAL: Adaptive Retrieval Loop for Culturally-Aligned Multilingual RAG

2026-04-28 · Nayeon Lee, Jiwoo Song, Byeongcheol Kang arxiv

Multilingual retrieval-augmented generation (mRAG) is often implemented within a fixed retrieval space, typically via query or document translation or multilingual embedding vector representations. However, this approach…

Cross-lingual Retrieval for Iterative Self-Supervised Training

2020-06-16 · NeurIPS 2020 12 · Chau Tran, Yuqing Tang, Xi-An Li, Jiatao Gu

Recent studies have demonstrated the cross-lingual alignment ability of multilingual pretrained language models. In this work, we found that the cross-lingual alignment can be further improved by training seq2seq models …

Machine Translation Retrieval Sentence Sentence Retrieval+2

Iterative Alignment Flows

2021-04-15 · Zeyu Zhou, Ziyu Gong, Pradeep Ravikumar, David I. Inouye

The unsupervised task of aligning two or more distributions in a shared latent space has many applications including fair representations, batch effect mitigation, and unsupervised domain adaptation. Existing flow-based …

Domain Adaptation Unsupervised Domain Adaptation

Stateful Evidence-Driven Retrieval-Augmented Generation with Iterative Reasoning

2026-03-25 · Qi Dong, Ziheng Lin, Ning Ding arxiv

Retrieval-Augmented Generation (RAG) grounds Large Language Models (LLMs) in external knowledge but often suffers from flat context representations and stateless retrieval, leading to unstable performance. We propose Sta…

Question Answering

When Iterative RAG Beats Ideal Evidence: A Diagnostic Study in Scientific Multi-hop Question Answering

2026-01-27 · Mahdi Astaraki, Mohammad Arshi Saloot, Ali Shiraee Kasmaee, Hamidreza Mahyar 외 arxiv

Retrieval-Augmented Generation (RAG) extends large language models (LLMs) beyond parametric knowledge, yet it is unclear when iterative retrieval-reasoning loops meaningfully outperform static RAG, particularly in scient…

Multi-hop Question Answering