Papers by Wei-Yu Kao
MAGIC: Multi-Argument Generation with Self-Refinement for Domain Generalization in Automatic Fact-Checking (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing methods for fact-checking are limited in retrieving evidence from documents . retrieved evidence derived from different sources strains generalization capabilities of classifiers . |
| Approach: | They propose a framework for cross-domain fact-checking using multi-argument generation . they propose to reconstruct concise evidence from large amounts of evidence retrieved from different sources . |
| Outcome: | The proposed framework is effective in identifying the veracity of out-of-domain claims . it can be used to extract evidence from documents and verify claims across domains . |