Variational Reasoning for Question Answering with Knowledge Graph

By Yuyu Zhang, Hanjun Dai, Zornitsa Kozareva, Alex Smola, Le Song. AAAI Conference on Artificial Intelligence, 2017

576 citations recorded by the source corpus.

Knowledge graph (KG) is known to be helpful for the task of question answering (QA), since it provides well-structured relational information between entities, and allows one to further infer indirect facts. However, it is challenging to build QA systems which can learn to reason over knowledge graphs based on question-answer pairs alone. First, when people ask questions, their expressions are noisy (for example, typos in texts, or variations in pronunciations), which is non-trivial for the QA system to match those mentioned entities to the knowledge graph. Second, many questions require multi-hop logic reasoning over the knowledge graph to retrieve the answers. To address these challenges, we propose a novel and unified deep learning architecture, and an end-to-end variational learning algorithm which can handle noise in questions, and learn multi-hop reasoning simultaneously. Our method achieves state-of-the-art performance on a recent benchmark dataset in the literature. We also derive a series of new benchmark datasets, including questions for multi-hop reasoning, questions paraphrased by neural translation model, and questions in human voice. Our method yields very promising results on all these challenging datasets.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/95280565aa3d120c6d7e8d87ea3423f16977f19a

How researchers see this paper

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

In Artificial Intelligence

Citations in the corpus

Reasoning on Graphs: Faithful and Interpretable Large Language Model Reasoning (2023, methodology): “We further select the MetaQA-3hop dataset (Zhang et al., 2018) which is based on Wiki-Movies KGs 6 .”
Question Answering Over Temporal Knowledge Graphs (2021, methodology): “However, similar to Zhang et al. (2017), we chose not to make this procedure a part of the dataset, to remove unwelcome dependence of QA systems on such formal candidate collection methods.”
Answering Visual-Relational Queries in Web-Extracted Knowledge Graphs (2017, background): “Another line of research is concerned with extensions of the link prediction problem to multi-hop reasoning [Zhang et al., 2018].”
Improving Multi-hop Knowledge Base Question Answering by Learning Intermediate Supervision Signals (2021, unknown): “rn the model. 5 EXPERIMENT In this section, we perform the evaluation experiments for our approach on the KBQA task. 5.1 Datasets We adopt three benchmark datasets for the multi-hop KBQA task: MetaQA [38] contains more than 400k single and multi-hop (up to 3-hop) questions in the domain of movie, co”
HybridQA: A Dataset of Multi-Hop Question Answering over Tabular and Textual Data (2020, background): “…of the XXXI Olympiad (Portuguese : Jogos da XXXI Olimpíada) and commonly known as Rio 2016 , was an international multi-sport event …… Be-rant et al., 2013), WebQSP (Yih et al., 2016), Web-Complex (Talmor and Berant, 2018), MetaQA (Zhang et al., 2018), WikiTableQuestion (Pasupat and Liang, 2015).”
DDRQA: Dynamic Document Reranking for Open-domain Multi-hop Question Answering (2020, background): “Graph neural networks (GNNs) have been shown effective on knowledge-based QA tasks by reasoning over graphs (De Cao et al., 2019; Sorokin and Gurevych, 2018; Zhang et al., 2018).”
A Survey on Knowledge Graphs: Representation, Acquisition, and Applications (2020, background): “ent answer generation, where external commonsense knowledge is utilized by relational path selection from ConceptNet and injection with selectivelygated attention. Variational Reasoning Network (VRN) [161] conducts multi-hop logic reasoning with reasoning-graph embedding, while handles the uncertain”
StructGPT: A General Framework for Large Language Model to Reason over Structured Data (2023, methodology): “For KG based QA (KGQA), we adopt two benchmark datasets, i.e., WebQuestionsSP (We-bQSP) (Yih et al., 2016) and MetaQA (Zhang et al., 2018) for evaluation.”
KG^2: Learning to Reason Science Exam Questions with Contextual Knowledge Graph Embeddings (2018, methodology): “Graph embedding has provided the representational ﬂexibility for neural models in many NLP tasks, such as dialog system [He et al., 2017], question answering [Zhang et al., 2017], link prediction [Bordes et al., 2013] and triple classiﬁcation [Feng et al., 2016].”
Improving Multi-hop Question Answering over Knowledge Graphs using Knowledge Base Embeddings (2020, methodology): “Question Answering over Knowledge Graphs (KGQA) has emerged as an important research area over the last few years (Zhang et al., 2018; Sun et al., 2019a).”
LEGO: Latent Execution-Guided Reasoning for Multi-Hop Question Answering on Knowledge Graphs (2021, methodology): “We evaluate L EGO on three large-scale multi-hop KGQA benchmark datasets: MetaQA (Zhang et al., 2018), WebQuestionsSP (WQSP) (Yih et al., 2015) and ComplexWebQuestions (CWQ) (Talmor & Berant, 2018).”
PullNet: Open Domain Question Answering with Iterative Retrieval on Knowledge Bases and Text (2019, methodology): “, 2018), and VRN (Zhang et al., 2018).”
Graph Retrieval-Augmented Generation: A Survey (2024, unknown): “For instance, Subgraph Retriever [181] adopts an alternating training paradigm, where the retriever’s parameters are fixed to use the graph data for training the generator.”
Embedding Logical Queries on Knowledge Graphs (2018, background): “Another well-studied thread in this space involves leveraging knowledge graphs to improve natural language question answering (QA) [4, 5, 47].”
Towards a Universal Continuous Knowledge Base (2020, background): “, 2017), natural language processing (Zhang et al., 2018), and bioinformatics (Ernst et al.”
Leveraging Domain Context for Question Answering Over Knowledge Graph (2019, background): “The difficulty issues lie at not only the vague question description but also a variety of entity and relationship types within the knowledge graph [23, 31].”
Answering Any-hop Open-domain Questions with Iterative Document Reranking (2020, background): “Recently, Graph neural networks (GNNs) have been shown effective on knowledge-based QA tasks by reasoning over graphs [6, 28, 42].”