List your company

Question Answering on SQuAD v2 (val)

83.72EMFLAN T5 + ALIGN Verifier
51.282459.703768.12576.5463Jul 6, 2023

Evaluation results

MethodLinks
FLAN T5 + ALIGN VerifierVerifier=ALIGN, Unanswerable threshold selection=SQuAD v2 validation split2023.0783.7286.550.95
FLAN T5mode=Zero-shot2023.0775.7279.010.83
GPT-3.5 + ALIGN VerifierVerifier=ALIGN, Unanswerable threshold selection=SQuAD v2 validation split2023.0767.1977.630.93
GPT-3.5mode=Zero-shot2023.0752.5363.960.76