mrm8488/spanbert-base-finetuned-tacred

Facebook Research fine-tuned SpanBERT on TACRED dataset by them . A large-scale relation extraction dataset with 106k+ examples over 42 TAC KBP relation types . The numbers marked as * are evaluated on the development sets because those models were not submitted to the official SQuAD leaderboard. All the other numbers are test numbers.Note: Numbers marked as ‘*’ are evaluated only on development sets. All of the numbers marked '*' are evaluated because they weren't submitted to SQuad leaderboard . The results were compared to those from the official leaderboard, which is based on the data from Facebook Research and the data was created by Manuel,