DeepAI AI Chat
Log In Sign Up

Inferential Text Generation with Multiple Knowledge Sources and Meta-Learning

by   Daya Guo, et al.

We study the problem of generating inferential texts of events for a variety of commonsense like if-else relations. Existing approaches typically use limited evidence from training examples and learn for each relation individually. In this work, we use multiple knowledge sources as fuels for the model. Existing commonsense knowledge bases like ConceptNet are dominated by taxonomic knowledge (e.g., isA and relatedTo relations), having a limited number of inferential knowledge. We use not only structured commonsense knowledge bases, but also natural language snippets from search-engine results. These sources are incorporated into a generative base model via key-value memory network. In addition, we introduce a meta-learning based multi-task learning algorithm. For each targeted commonsense relation, we regard the learning of examples from other relations as the meta-training process, and the evaluation on examples from the targeted relation as the meta-test process. We conduct experiments on Event2Mind and ATOMIC datasets. Results show that both the integration of multiple knowledge sources and the use of the meta-learning algorithm improve the performance.


page 1

page 2

page 3

page 4


Understanding Few-Shot Commonsense Knowledge Models

Providing natural language processing systems with commonsense knowledge...

Generative Meta-Learning for Zero-Shot Relation Triplet Extraction

The zero-shot relation triplet extraction (ZeroRTE) task aims to extract...

Mapping and Cleaning Open Commonsense Knowledge Bases with Generative Translation

Structured knowledge bases (KBs) are the backbone of many know­ledge-int...

KEML: A Knowledge-Enriched Meta-Learning Framework for Lexical Relation Classification

Lexical relations describe how concepts are semantically related, in the...

Natural Language to Structured Query Generation via Meta-Learning

In conventional supervised training, a model is trained to fit all the t...

Commonsense Knowledge in Wikidata

Wikidata and Wikipedia have been proven useful for reason-ing in natural...

kogito: A Commonsense Knowledge Inference Toolkit

In this paper, we present kogito, an open-source tool for generating com...