Papers by Balakrishnan Narayanaswamy
What we need to learn if we want to do and not just talk (N18-3)
Copied to clipboard
| Challenge: | Existing methods for task-oriented dialogs require fluent natural language responses and correct external actions . but they perform poorly in real world dialog tasks, a new study shows . |
| Approach: | They propose a hybrid model where nearest neighbor is used to generate fluent responses and Seq2Seq type models ensure dialogue coherency and generate accurate external actions. |
| Outcome: | The proposed model achieves a 78% relative improvement in fluency and 200% improvement in accuracy of external calls. |
Recursive Template-based Frame Generation for Task Oriented Dialog (2020.acl-main)
Copied to clipboard
| Challenge: | The input of an NLU component is a semantic frame that captures the intent and slot-labels provided by the user. |
| Approach: | They propose a recursive, hierarchical representation that captures the intent and slot-labels provided by the user and extend local tree-based loss functions with terms that provide global supervision. |
| Outcome: | The proposed representation improves on the widely used ATIS dataset and significantly improves the performance of the proposed framework. |
Joint Multiple Intent Detection and Slot Labeling for Goal-Oriented Dialog (N19-1)
Copied to clipboard
| Challenge: | Neural network models have gained traction for sentence-level intent classification and token-based slot-label identification. |
| Approach: | They propose a neural network model that performs multi-label classification for identifying multiple intents and produces token-based slot-l labels at the token-level. |
| Outcome: | The proposed model provides a small but statistically significant improvement on the ATIS dataset and 55% accuracy improvement on an internal multi-intent dataset. |