Papers by Ayesha Enayet
An Analysis of Dialogue Act Sequence Similarity Across Multiple Domains (2022.lrec-1)
Copied to clipboard
| Challenge: | a recent study shows that many machine learning models perform poorly when exposed to domain shifts due to contextual differences. |
| Approach: | They analyze dialogue act sequences from related domains to predict performance degradation . they find that when dialogue acts sequences are dissimilar they lie further away in embedding space . |
| Outcome: | The proposed model can be trained even when the datasets are corrupted with noise. |
Improving the Generalizability of Collaborative Dialogue Analysis With Multi-Feature Embeddings (2023.eacl-main)
Copied to clipboard
| Challenge: | Using a multi-feature embedding improves the generalizability of conflict prediction models trained on dialogues. |
| Approach: | They propose a multi-feature embedding that leverages textual, structural, and semantic information from dialogues by incorporating lexical, dialogue acts, and sentiment features. |
| Outcome: | The proposed model is excellent domain-agnostic representation for meta-pretraining a few-shot model on collaborative multiparty dialogues. |