Retrieving and Reading: A Comprehensive Survey on Open-domain Question Answering (2101.00774v3)

Abstract: Open-domain Question Answering (OpenQA) is an important task in NLP, which aims to answer a question in the form of natural language based on large-scale unstructured documents. Recently, there has been a surge in the amount of research literature on OpenQA, particularly on techniques that integrate with neural Machine Reading Comprehension (MRC). While these research works have advanced performance to new heights on benchmark datasets, they have been rarely covered in existing surveys on QA systems. In this work, we review the latest research trends in OpenQA, with particular attention to systems that incorporate neural MRC techniques. Specifically, we begin with revisiting the origin and development of OpenQA systems. We then introduce modern OpenQA architecture named "Retriever-Reader" and analyze the various systems that follow this architecture as well as the specific techniques adopted in each of the components. We then discuss key challenges to developing OpenQA systems and offer an analysis of benchmarks that are commonly used. We hope our work would enable researchers to be informed of the recent advancement and also the open challenges in OpenQA research, so as to stimulate further progress in this field.

Overview of "Bare Demo of IEEEtran.cls for IEEE Computer Society Journals"

The document in question is a demo file that serves as a template for authors submitting papers to IEEE Computer Society journals using the IEEEtran.cls style file with \LaTeX. This paper does not delve into any specific research findings but instead provides a structured format intended for the production of academic papers within the IEEE community.

Purpose and Utility

The primary goal of this template is to assist authors in effectively formatting their papers according to the IEEE Computer Society requirements. By using this starter file, authors can focus on their scientific content while ensuring that the document adheres to editorial and publication standards necessary for submission.

Structure and Components

The paper exemplifies the essential components of a typical IEEE journal paper:

• Title and Author Information: It includes the correct placement of the title, author names, and the respective affiliations.
• Abstract and Keywords: The template provides placeholders for an abstract and a list of keywords that are crucial for summarizing the paper's content and for indexing purposes.
• Sections and Headings: The document demonstrates the formatting of various section levels, from primary sections to subsections and subsubsections, aligning with IEEE's stylistic guidelines.
• Appendices and References: Guidance on appending additional material and citing literature works is provided to maintain citation integrity and transparency.
• Acknowledgements and Biographies: It offers sections to recognize contributions from non-authors and provides biographical sketches of the researchers.

Theoretical and Practical Implications

While the template itself does not contribute innovative theories or methodologies, its implications lie in standardization. Researchers benefit from a cohesive layout that mitigates formatting discrepancies across submissions, thus emphasizing content over presentation issues. The practical advantage is a reduction in the time and effort expended on formatting, facilitating a quicker review process and enhancing the focus on the research contributions being presented.

Future Developments

Looking forward, an area of potential development is the integration of this template with collaborative platforms or automated systems to streamline the submission process further. As collaborative research environments evolve, enhancements to such templates could include automatic referencing updates, pre-submission checks for common formatting errors, and compatibility with a broader range of collaborative and typesetting tools.

The significance of maintaining and updating these templates grows with the internationalization and cross-disciplinarity of research collaborations, ensuring that a wide audience can adhere to rigorous and uniform standards of academic publishing.