Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2405.15585 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912075763154944 |
|---|---|
| author | Saley, Vishal Vivek Das, Rocktim Jyoti Raghu, Dinesh Mausam |
| author_facet | Saley, Vishal Vivek Das, Rocktim Jyoti Raghu, Dinesh Mausam |
| contents | End-to-end Task-Oriented Dialog (TOD) systems typically require extensive training datasets to perform well. In contrast, large language model (LLM) based TOD systems can excel even with limited data due to their ability to learn tasks through in-context exemplars. However, these models lack alignment with the style of responses in training data and often generate comprehensive responses, making it difficult for users to grasp the information quickly. In response, we propose SyncTOD that synergizes LLMs with task-specific hints to improve alignment in low-data settings. SyncTOD employs small auxiliary models to provide hints and select exemplars for in-context prompts. With ChatGPT, SyncTOD achieves superior performance compared to LLM-based baselines and SoTA models in low-data settings, while retaining competitive performance in full-data settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_15585 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Synergizing In-context Learning with Hints for End-to-end Task-oriented Dialog Systems Saley, Vishal Vivek Das, Rocktim Jyoti Raghu, Dinesh Mausam Computation and Language End-to-end Task-Oriented Dialog (TOD) systems typically require extensive training datasets to perform well. In contrast, large language model (LLM) based TOD systems can excel even with limited data due to their ability to learn tasks through in-context exemplars. However, these models lack alignment with the style of responses in training data and often generate comprehensive responses, making it difficult for users to grasp the information quickly. In response, we propose SyncTOD that synergizes LLMs with task-specific hints to improve alignment in low-data settings. SyncTOD employs small auxiliary models to provide hints and select exemplars for in-context prompts. With ChatGPT, SyncTOD achieves superior performance compared to LLM-based baselines and SoTA models in low-data settings, while retaining competitive performance in full-data settings. |
| title | Synergizing In-context Learning with Hints for End-to-end Task-oriented Dialog Systems |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2405.15585 |