Papers by Jun Shern Chan
Few-shot Adaptation Works with UnpredicTable Data (2023.acl-long)
Copied to clipboard
| Challenge: | Prior work on language models (LMs) shows that training on a large number of diverse tasks improves few-shot learning (FSL) performance on new tasks. |
| Approach: | They finetuned 413,299 tasks from internet tables to find narrow subsets outperform more diverse datasets. |
| Outcome: | The proposed model outperforms training on 40 human-curated NLP datasets on 52 downstream tasks, but not proportionally to dataset scale. |