[Adapt] Language Generation with Multi-Hop Reasoning on Commonsense Knowledge Graph

[Li Zitong]

Hi Adapters,

This time, I will introduce a paper accepted by EMNLP2020: Language Generation with Multi-Hop Reasoning on Commonsense Knowledge Graph.

Despite the success of generative pre-trained language models on a series of text generation tasks, they still suffer in cases where reasoning over underlying commonsense knowledge is required during generation. The writers argue that exploiting both the structural and semantic information of the knowledge graph facilitates commonsenseaware text generation. In this paper, the writers propose Generation with Multi-Hop Reasoning Flow (GRF) that enables pre-trained models with dynamic multi-hop reasoning on multirelational paths extracted from the external commonsense knowledge graph. They empirically show that their model outperforms existing baselines on three text generation tasks that require reasoning over commonsense knowledge. They also demonstrate the effectiveness of the dynamic multi-hop reasoning module with reasoning paths inferred by the model that provide rationale to the generation.


Time: Wed 4:00pm

Venue: SEIEE 3-414


Zitong

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://cs.sjtu.edu.cn/pipermail/adapt/attachments/20201201/04029c86/attachment.html>