Reading Comprehension tests are commonly used to assess the degree to which people comprehend what they read. This is why we work with the hypothesis that it is reasonable to use these tests to assess the degree to which a machine “comprehends” what it is reading. In this work, we evaluate Question Answering systems using Reading Comprehension tests from exams to enter University. This article analyses the datasets generated, the kind of inferences required, the methodology followed in three evaluation campaigns, the approaches presented by participants and current results. Besides, we study the evolution of systems and the main lessons learned in this evaluation process. We also show how current technologies are unable to pass university-entrance exams. This is because these tests require a deep understanding of texts, as well as detecting the similar meaning of phrases with different words. Future directions focused on these ideas seem more promising than including a massive amount of data for training systems, what has allowed systems to obtain outstanding results in Reading Comprehension tests with more straightforward questions. We think this study helps to increase the knowledge about how to develop better Question Answering systems. 相似文献
In order to investigate the situation of political cognition and ability, the research group had surveyed some rural areas by the methods of questionnaires, literature review and case interviews. The d... 相似文献
Purpose: Approaches to build farmers' analytical capacities are said to trigger wide-ranging changes. This article reports on the communication process between participants and non-participants in one such approach, related to the technical and management skills learned by participants and the changes these participants subsequently made, and the outcomes in terms of non-participants' learning.
Design/Methodology/Approach: In this study, we analysed the following: (1) participants' learning and changes in social practices; (2) communication between participants and non-participants regarding technical and management skills and changes in social practices; (3) non-participants' learning and changes in social practices. The case study was a management advice to family farm project in three villages in Benin.
Findings: Most participants learned management skills, which led them to reassess their objectives and to reduce traditional social practices they now considered unproductive. Even in the case of frequent communication, non-participants found it difficult to learn management skills, which hindered their experiential learning. Non-participants consequently had difficulty understanding why participants changed their social practices such as reduction of their traditional gift giving, leading to limited well-argued discussion about these changes in practices between participants and non-participants.
Practical Implications: This study shows that, due to the limited learning process of non-participants communicating with participants, there is a need to design and test approaches to achieve capacity-building while including more participants at a similar cost, and to stimulate explicit discussion at village level about the changes in values and social practices that these approaches may trigger.
Originality/Value: The article analyses both participants and non-participants' experiential learning, and looks at the two sides of the interaction between the communication processes and non-participants' learning. 相似文献