A Self-Training Method for Machine Reading Comprehension with Soft Evidence Extraction

May 11, 2020 · Entered Twilight · 🏛 Annual Meeting of the Association for Computational Linguistics

"Last commit was 5.0 years ago (≥5 year threshold)"

Evidence collected by the PWNC Scanner

Repo contents: .gitignore, README.md, bert_model, combine_label.py, data, data_preprocess, ensemble_boolq.py, ensemble_coqa.py, eval_utils, experiments, extract_evidence.py, general_util, inter_label.py, label_main.py, main.py, main2_0.6.2_topk.py, main2_0.6.2_topk_predict_sentences.py, main2_multi_choice_top_k_evidence.py, main_0.6.2.py, main_0.6.2_multiple.py, main_0.6.2_topk.py, main_0.6.2_topk_predict_sentences.py, main_0.6.2_topk_sent_pretrain.py, main_1.0.py, main_multi_choice.py, main_multi_choice_multiple_evidence.py, main_multi_choice_top_k_evidence.py, main_multi_choice_top_k_evidence_amp.py, main_x2.py, merge.py, predict_sentence_main0.6.2.py, predict_sentence_main2_0.6.2.py, pretrain_attention_main.py, reader, requirements.txt, scripts, simple_main.py, test_label_acc.py, top_k.py, transfer_main.py, union_label.py

Authors Yilin Niu, Fangkai Jiao, Mantong Zhou, Ting Yao, Jingfang Xu, Minlie Huang arXiv ID 2005.05189 Category cs.CL: Computation & Language Citations 36 Venue Annual Meeting of the Association for Computational Linguistics Repository https://github.com/SparkJiao/Self-Training-MRC ⭐ 34 Last Checked 1 month ago

Abstract

Neural models have achieved great success on machine reading comprehension (MRC), many of which typically consist of two components: an evidence extractor and an answer predictor. The former seeks the most relevant information from a reference text, while the latter is to locate or generate answers from the extracted evidence. Despite the importance of evidence labels for training the evidence extractor, they are not cheaply accessible, particularly in many non-extractive MRC tasks such as YES/NO question answering and multi-choice MRC. To address this problem, we present a Self-Training method (STM), which supervises the evidence extractor with auto-generated evidence labels in an iterative process. At each iteration, a base MRC model is trained with golden answers and noisy evidence labels. The trained model will predict pseudo evidence labels as extra supervision in the next iteration. We evaluate STM on seven datasets over three MRC tasks. Experimental results demonstrate the improvement on existing MRC models, and we also analyze how and why such a self-training method works in MRC. The source code can be obtained from https://github.com/SparkJiao/Self-Training-MRC