Hi HuggingFace Datasets team!
VORTEXRAG was evaluated on 6 datasets available on the HuggingFace Hub using datasets.load_dataset():
- NQ, TriviaQA, WebQ, PopQA, HotpotQA, 2WikiMultiHopQA
Results: EM 74.8, F1 82.6 averaged across all 6. Full evaluation code is in the repo.
Hi HuggingFace Datasets team!
VORTEXRAG was evaluated on 6 datasets available on the HuggingFace Hub using datasets.load_dataset():
Results: EM 74.8, F1 82.6 averaged across all 6. Full evaluation code is in the repo.