Question Answering through Transfer Learning from Large Fine-grained Supervision Data
arXiv:1702.02171
Abstract
We show that the task of question answering (QA) can significantly benefit from the transfer learning of models trained on a different large, fine-grained QA dataset. We achieve the state of the art in two well-studied QA datasets, WikiQA and SemEval-2016 (Task 3A), through a basic transfer learning technique from SQuAD. For WikiQA, our model outperforms the previous best model by more than 8%. We demonstrate that finer supervision provides better guidance for learning lexical and syntactic information than coarser supervision, through quantitative results and visual analysis. We also show that a similar transfer learning procedure achieves the state of the art on an entailment task.
Published as a conference paper at ACL 2017 (short paper). Code available at https://github.com/shmsw25/qa-transfer
References in corpus (6)
- ADADELTA: An Adaptive Learning Rate Method
- Bidirectional Attention Flow for Machine Comprehension
- Dynamic Coattention Networks For Question Answering
- Machine Comprehension Using Match-LSTM and Answer Pointer
- Bilateral Multi-Perspective Matching for Natural Language Sentences
- NewsQA: A Machine Comprehension Dataset
Cited by in corpus (8)
- Few-shot Learning for Named Entity Recognition in Medical Text
- Adversarial TableQA: Attention Supervision for Question Answering on Tables
- Finding Answers from the Word of God: Domain Adaptation for Neural Networks in Biblical Question Answering
- Cross-Lingual Transfer Learning for Question Answering
- AdaFilter: Adaptive Filter Fine-tuning for Deep Transfer Learning
- Unsupervised Domain Adaptation on Reading Comprehension
- Reasoning-Driven Question-Answering for Natural Language Understanding
- SNU_IDS at SemEval-2018 Task 12: Sentence Encoder with Contextualized Vectors for Argument Reasoning Comprehension