Rethinking Loss Functions for Fact Verification (EACL) Rethinking Loss Functions for Fact Verification (EACL)

Yuta Mukobara*, Yutaro Shigeto**, and Masashi Shimbo.
Rethinking Loss Functions for Fact Verification.
The 18th Conference of the European Chapter of the Association for Computational Linguistics (EACL), pp. 432-442, March 2024.

* Work conducted during an internship at STAIR Lab
** Corresponding Author

We explore loss functions for fact verification in the FEVER shared task. While the cross-entropy loss is a standard objective for training verdict predictors, it fails to capture the heterogeneity among the FEVER verdict classes. In this paper, we develop two task-specific objectives tailored to FEVER. Experimental results confirm that the proposed objective functions outperform the standard cross-entropy. Performance is further improved when these objectives are combined with simple class weighting, which effectively overcomes the imbalance in the training data.

[ ACL Anthology | arXiv | Code ]