Rethinking Ground Truth: A Case Study on Human Label Variation in MLLM Benchmarking

Tomás Ruiz, Tanalp Agustoslu, Carsten Schwemmer. Rethinking Ground Truth: A Case Study on Human Label Variation in MLLM Benchmarking. In IEEE International Conference on Big Data, BigData 2025, Macau, China, December 8-11, 2025. pages 5224-5229, IEEE, 2025. [doi]

Abstract

Abstract is missing.