Can Generative Pre-trained Language Models Serve as Knowledge Bases for Closed-book QA?
About
Recent work has investigated the interesting question using pre-trained language models (PLMs) as knowledge bases for answering open questions. However, existing work is limited in using small benchmarks with high test-train overlaps. We construct a new dataset of closed-book QA using SQuAD, and investigate the performance of BART. Experiments show that it is challenging for BART to remember training facts in high precision, and also challenging to answer closed-book questions even if relevant knowledge is retained. Some promising directions are found, including decoupling the knowledge memorizing process and the QA finetune process, forcing the model to recall relevant knowledge when question answering.
Cunxiang Wang, Pai Liu, Yue Zhang• 2021
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Closed-book Question Answering | TriviaQA TQ (test) | -- | 9 | |
| Closed-book Question Answering | NaturalQuestions (test) | -- | 9 | |
| Question Answering | SQuAD 20 passages subset 1.1 (test) | RA0.873 | 5 | |
| Question Answering | SQuAD 160 passages subset 1.1 (test) | RA79.6 | 5 | |
| Question Answering | SQuAD 547 passages 1.1 (test) | RA66.3 | 5 | |
| Closed-book Question Answering | SQuAD adaptation 2 (test) | EM1.8 | 2 | |
| Closed-book Question Answering | WebQuestions WB (test) | -- | 1 |
Showing 7 of 7 rows