Resources

• Knowledge-Constrained Generation of Implicit Information between Sentences
Code for Equipping Language Models with Commonsense Knowledge for Generating Implicit Knowledge Statements on Github
• CO-NNECT: A Framework for Establishing Single- and Multihop Commonsense Knowledge Paths between Concepts in Texts
Code for CO-NNECT on Github
• CoCo-Ex: A Tool for Extracting Concepts from Texts and Linking them to ConceptNet
Code for CoCO-Ex on Github (Becker et al., 2021)
• Commonsense Knowledge Relation Classification System (COREC)
Code for our Commonsense Knowledge Relation Classifier on Github (Becker et al., 2019)
• Commonsense Knowledge Relation Classification with pretrained Langauge Models (COREC-LM)
Code for our Enhanced Version of COREC (COREC-LM) on Github
• Semantic Clause Types Classifier
Code for our Semantic Clause Types Classifier on Github (Becker et al., 2017)
• IKAT-DE: Implied knowledge in argumentative texts (German Version)
Dataset which contains annotations about implicit information in argumentative texts in German (Becker et al., 2017)
• IKAT-EN: Implied knowledge in argumentative texts (English Version)
Dataset which contains annotations about implicit information in argumentative texts in English (Becker et al., 2020)
• Situation Entity Type labelled corpus for German (GER_SET)
Corpus which contains German texts from different genres annotated with Semantic Clause Types (e.g. States, Events, Generic Sentences...) (Becker et al., 2017)