Takao Morita

Look Before You Train: Train-Test Contamination in GlobalQA and a Training-Free Method that Reverse-Engineers Its Gold Rules

GlobalQA is a recent benchmark for corpus-level question answering (counting, sorting, and extremum queries over a resume corpus), on which RL-based retrieval agents such as MARAG-R1 report a state-of-the-art answer-F1 of 28.60-31.22. While calibrating a simple training-free syst...

Look Before You Train: Train-Test Contamination in GlobalQA and a Training-Free Method that Reverse-Engineers Its Gold Rules

GlobalQA is a recent benchmark for corpus-level question answering (counting, sorting, and extremum queries over a resume corpus), on which RL-based retrieval agents such as MARAG-R1 report a state-of-the-art answer-F1 of 28.60-31.22. While calibrating a simple training-free syst...