Papers by Yonggang Wen

2 papers
Are Missing Links Predictable? An Inferential Benchmark for Knowledge Graph Completion (2021.acl-long)

Copied to clipboard

Challenge: Existing benchmarks for Knowledge Graph Completion (KGC) are unsatisfactory .
Approach: They propose to use rule-guided train/test generation instead of conventional random split to ensure that each testing sample is predictable with supportive data in the training set.
Outcome: The proposed model improves on existing benchmarks in inferential ability, assumptions, and patterns.
Discriminative Reasoning with Sparse Event Representation for Document-level Event-Event Relation Extraction (2023.acl-long)

Copied to clipboard

Challenge: Document-level Event Causality Identification (DECI) is a sentence-level task that requires long-text understanding.
Approach: They propose a document-level event causality identification model (SENDIR) that uses sparse attention to capture long-distance dependence.
Outcome: The proposed model can be used to discriminate between event pairs in the same sentence or span multiple sentences.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations