Work

Rethinking Loss Functions for Fact Verification

Year: 2024

Type: preprint

Abstract: 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. The souce code is available at https://github.com/yuta-mukobara/RLF-KGAT

Source: arXiv (Cornell University)

DOI: 10.48550/arxiv.2403.08174

Authors: Mukobara, Yuta, Shigeto, Yutaro, Shimbo, Masashi

Institutions: Tokyo Institute of Technology, RIKEN, RIKEN Center for Advanced Intelligence Project, Chiba Institute of Technology

Language: English

Topic: Topic Modeling

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funder: Japan Society for the Promotion of Science

Award: Development of a Zero-Shot Caption Generation Method

OpenAlex ID: W4392822340

Cites:

Cited by:

Related to: 10