We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:


References & Citations

DBLP - CS Bibliography


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computation and Language

Title: Bridging the Gap between Language Model and Reading Comprehension: Unsupervised MRC via Self-Supervision

Abstract: Despite recent success in machine reading comprehension (MRC), learning high-quality MRC models still requires large-scale labeled training data, even using strong pre-trained language models (PLMs). The pre-training tasks for PLMs are not question-answering or MRC-based tasks, making existing PLMs unable to be directly used for unsupervised MRC. Specifically, MRC aims to spot an accurate answer span from the given document, but PLMs focus on token filling in sentences. In this paper, we propose a new framework for unsupervised MRC. Firstly, we propose to learn to spot answer spans in documents via self-supervised learning, by designing a self-supervision pretext task for MRC - Spotting-MLM. Solving this task requires capturing deep interactions between sentences in documents. Secondly, we apply a simple sentence rewriting strategy in the inference stage to alleviate the expression mismatch between questions and documents. Experiments show that our method achieves a new state-of-the-art performance for unsupervised MRC.
Subjects: Computation and Language (cs.CL)
Cite as: arXiv:2107.08582 [cs.CL]
  (or arXiv:2107.08582v1 [cs.CL] for this version)

Submission history

From: Ning Bian [view email]
[v1] Mon, 19 Jul 2021 02:14:36 GMT (139kb,D)

Link back to: arXiv, form interface, contact.