Award-Winning Research on Social Bias in QA Benchmarks

2025-12-20
1 min read

Congratulations to Angelie Kraft, Judith Simon, and Sonja Schimmler for receiving an “Honorable Mention” at the 14th International Joint Conference on Natural Language Processing (IJCNLP-AACL 2025).

Their paper, “Social Bias in Popular Question-Answering Benchmarks,” investigates how social biases infiltrate AI benchmarks. Their findings emphasize that building transparent, diversity-sensitive datasets is essential for responsible AI development.

Find the paper here.