LEGO: Latent Execution-Guided Reasoning for Multi-Hop Question Answering on Knowledge Graphs

Hongyu Ren, Hanjun Dai, Bo Dai, Xinyun Chen, Michihiro Yasunaga, Haitian Sun, Dale Schuurmans, Jure Leskovec, Denny Zhou
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:8959-8970, 2021.

Abstract

Answering complex natural language questions on knowledge graphs (KGQA) is a challenging task. It requires reasoning with the input natural language questions as well as a massive, incomplete heterogeneous KG. Prior methods obtain an abstract structured query graph/tree from the input question and traverse the KG for answers following the query tree. However, they inherently cannot deal with missing links in the KG. Here we present LEGO, a Latent Execution-Guided reasOning framework to handle this challenge in KGQA. LEGO works in an iterative way, which alternates between (1) a Query Synthesizer, which synthesizes a reasoning action and grows the query tree step-by-step, and (2) a Latent Space Executor that executes the reasoning action in the latent embedding space to combat against the missing information in KG. To learn the synthesizer without step-wise supervision, we design a generic latent execution guided bottom-up search procedure to find good execution traces efficiently in the vast query space. Experimental results on several KGQA benchmarks demonstrate the effectiveness of our framework compared with previous state of the art.

Cite this Paper

BibTeX
@InProceedings{pmlr-v139-ren21a, title = {LEGO: Latent Execution-Guided Reasoning for Multi-Hop Question Answering on Knowledge Graphs}, author = {Ren, Hongyu and Dai, Hanjun and Dai, Bo and Chen, Xinyun and Yasunaga, Michihiro and Sun, Haitian and Schuurmans, Dale and Leskovec, Jure and Zhou, Denny}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, pages = {8959--8970}, year = {2021}, editor = {Meila, Marina and Zhang, Tong}, volume = {139}, series = {Proceedings of Machine Learning Research}, month = {18--24 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v139/ren21a/ren21a.pdf}, url = {https://proceedings.mlr.press/v139/ren21a.html}, abstract = {Answering complex natural language questions on knowledge graphs (KGQA) is a challenging task. It requires reasoning with the input natural language questions as well as a massive, incomplete heterogeneous KG. Prior methods obtain an abstract structured query graph/tree from the input question and traverse the KG for answers following the query tree. However, they inherently cannot deal with missing links in the KG. Here we present LEGO, a Latent Execution-Guided reasOning framework to handle this challenge in KGQA. LEGO works in an iterative way, which alternates between (1) a Query Synthesizer, which synthesizes a reasoning action and grows the query tree step-by-step, and (2) a Latent Space Executor that executes the reasoning action in the latent embedding space to combat against the missing information in KG. To learn the synthesizer without step-wise supervision, we design a generic latent execution guided bottom-up search procedure to find good execution traces efficiently in the vast query space. Experimental results on several KGQA benchmarks demonstrate the effectiveness of our framework compared with previous state of the art.} }
Endnote
%0 Conference Paper %T LEGO: Latent Execution-Guided Reasoning for Multi-Hop Question Answering on Knowledge Graphs %A Hongyu Ren %A Hanjun Dai %A Bo Dai %A Xinyun Chen %A Michihiro Yasunaga %A Haitian Sun %A Dale Schuurmans %A Jure Leskovec %A Denny Zhou %B Proceedings of the 38th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Marina Meila %E Tong Zhang %F pmlr-v139-ren21a %I PMLR %P 8959--8970 %U https://proceedings.mlr.press/v139/ren21a.html %V 139 %X Answering complex natural language questions on knowledge graphs (KGQA) is a challenging task. It requires reasoning with the input natural language questions as well as a massive, incomplete heterogeneous KG. Prior methods obtain an abstract structured query graph/tree from the input question and traverse the KG for answers following the query tree. However, they inherently cannot deal with missing links in the KG. Here we present LEGO, a Latent Execution-Guided reasOning framework to handle this challenge in KGQA. LEGO works in an iterative way, which alternates between (1) a Query Synthesizer, which synthesizes a reasoning action and grows the query tree step-by-step, and (2) a Latent Space Executor that executes the reasoning action in the latent embedding space to combat against the missing information in KG. To learn the synthesizer without step-wise supervision, we design a generic latent execution guided bottom-up search procedure to find good execution traces efficiently in the vast query space. Experimental results on several KGQA benchmarks demonstrate the effectiveness of our framework compared with previous state of the art.
APA
Ren, H., Dai, H., Dai, B., Chen, X., Yasunaga, M., Sun, H., Schuurmans, D., Leskovec, J. & Zhou, D.. (2021). LEGO: Latent Execution-Guided Reasoning for Multi-Hop Question Answering on Knowledge Graphs. Proceedings of the 38th International Conference on Machine Learning, in Proceedings of Machine Learning Research 139:8959-8970 Available from https://proceedings.mlr.press/v139/ren21a.html.

Related Material