Papers with test-time augmentation
Test-time Augmentation for Factual Probing (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods to improve factual probing are relation-specific and do not generalize to unseen relation types. |
| Approach: | They propose to use test-time augmentation to augment and ensemble prompts at test time to reduce sensitivity to prompt variations. |
| Outcome: | The proposed method improves model confidence, but for other models, it leads to degradation. |
Test-Time Code-Switching for Cross-lingual Aspect Sentiment Triplet Extraction (2025.naacl-long)
Copied to clipboard
| Challenge: | Aspect Sentiment Triplet Extraction (ASTE) is a thriving research area . current code-switching methods suffer from term boundary detection issues and out-of-dictionary problems. |
| Approach: | They propose a test-time code-switching framework which bridges the gap between bilingual training and monolingual test- time prediction. |
| Outcome: | The proposed framework achieves an average improvement of 3.7% on four cross-lingual datasets. |
Mind Your Format: Towards Consistent Evaluation of In-Context Learning Improvements (2024.findings-acl)
Copied to clipboard
| Challenge: | Large language models demonstrate remarkable ability for learning to solve new tasks from a few examples. |
| Approach: | They propose to use templates to aggregate model predictions across multiple templates to improve model performance. |
| Outcome: | The proposed model ensembles boost model predictions while being robust to the choice of random set of templates. |