Papers by Xuemeng Song
Multi-source Semantic Graph-based Multimodal Sarcasm Explanation Generation (2023.acl-long)
Copied to clipboard
| Challenge: | Existing studies can only identify sarcastic post but could not give concrete explanation for why it is sarkastic. |
| Approach: | They propose a multimodal sarcasm explanation scheme that generates a sentence for a social post to explain why it contains sarkasis. |
| Outcome: | The proposed model outperforms existing methods on a public dataset. |
MERIt: Meta-Path Guided Contrastive Learning for Logical Reasoning (2022.findings-acl)
Copied to clipboard
| Challenge: | Existing methods to infer logical relations with annotated training data suffer from over-fitting and poor generalization problems due to the dataset sparsity. |
| Approach: | They propose a MEta-path guided contrastive learning method for logical ReasonIng of text that performs self-supervised pre-training on abundant unlabeled text data. |
| Outcome: | The proposed method outperforms the baselines on two logical reasoning benchmarks with significant improvements. |