MCScript2.0: A Machine Comprehension Corpus Focused on Script Events and Participants
44
Citation
37
Reference
10
Related Paper
Citation Trend
Abstract:
We introduce MCScript2.0, a machine comprehension corpus for the end-to-end evaluation of script knowledge. MCScript2.0 contains approx. 20,000 questions on approx. 3,500 texts, crowdsourced based on a new collection process that results in challenging questions. Half of the questions cannot be answered from the reading texts, but require the use of commonsense and, in particular, script knowledge. We give a thorough analysis of our corpus and show that while the task is not challenging to humans, existing machine comprehension models fail to perform well on the data, even if they make use of a commonsense knowledge base. The dataset is available at http://www.sfb1102. uni-saarland.de/?page_id=2582Keywords:
Commonsense knowledge
A comparative study of four reading journals' contributions to comprehension instruction methodology
Abstract This study analyzed and compared the content of four reading journals for the years 1973 and 1983 on eight categories of comprehension instruction and three categories of comprehension follow up. Of specific interest was the extent to which articles concretely described strategies for teaching comprehension or for following up on comprehension instruction according to specified definitions for each category. In general, results indicated an increase in the number of articles focusing on comprehension issues and practices.
Reciprocal teaching
Cite
Citations (1)
Commonsense knowledge is fundamental to make machines reach human-level intelligence. However, conventional methods of commonsense extraction generally do not work well because commonsense by nature is usually not explicitly stated in texts or other data. Besides, commonsense knowledge graphs built in advance are difficult to cover all the knowledge required for practical tasks due to the incompleteness of knowledge graphs. In this paper, we propose an online commonsense oracle to achieve knowledge reasoning. Specifically, we focus on the on-demand inference of specific commonsense propositions. We use capableOf relation as an example due to its notable significance in daily life. For more effective capableOf reasoning, informative supporting features derived from an existing commonsense knowledge graph and a Web search engine are exploited. Finally, we conduct extensive experiments, and the results demonstrate the effectiveness of our approach.
Commonsense knowledge
Commonsense reasoning
Knowledge graph
Cite
Citations (1)
Reading Comprehension is an illusive entity. It is difficult to define, measure and teach. Researchers attempting to construct a theoretical model of the comprehension process for empirical examination are often confronted with a maze of confusing studies. The untangled maze is perhaps more discouraging. Confusion generally gives way to contradiction. Measurement problems are also prevalent in the area of comprehension. Naturally, any quality which is difficult to isolate and describe is equally difficult to measure. Finally, teachers must face the ominous responsibility of helping students obtain a quality which is vaguely defined and measured. It is the most difficult of these dilemmas with which this discussion deals; teaching students to gain meaning from the printed page. TEACHING LANGUAGE CLUES TO READING COMPREHENSION
Contradiction
Confusion
Reciprocal teaching
Cite
Citations (1)
Reciprocal teaching
Cite
Citations (103)
Knowledge facts are typically represented by relational triples, while we observe that some commonsense facts are represented by the triples whose forms are inconsistent with the expression of language. This inconsistency puts forward a challenge for pre-trained language models to deal with these commonsense knowledge facts. In this paper, we term such knowledge as deep commonsense knowledge and conduct extensive exploratory experiments on it. We show that deep commonsense knowledge occupies a significant part of commonsense knowledge while conventional methods fail to capture it effectively. We further propose a novel method to mine the deep commonsense knowledge distributed in sentences, alleviating the reliance of conventional methods on the triple representation form of knowledge. Experiments demonstrate that the proposal significantly improves the performance in mining deep commonsense knowledge.
Commonsense knowledge
Commonsense reasoning
Representation
Cite
Citations (0)
Classroom teachers, in K–12 and even in college, can help students achieve higher levels of reading comprehension. Regardless of the content area, teachers can help struggling readers using the comprehension mini-lesson—a step-by-step research-based intervention.
Cite
Citations (0)
Reading in the digital world has both similarities and differences from reading on paper. Books as tangible objects elicit powerful responses linked to the pleasures felt in reading them. Although our eyes scan differently when reading online, reading e-versions of books initially seems similar to reading on paper. However digital books have some significantly different aspects that will be especially powerful in academic work.
Cite
Citations (0)
Commonsense knowledge bases such as ConceptNet represent knowledge in the form of relational triples. Inspired by recent work by Li et al., we analyse if knowledge base completion models can be used to mine commonsense knowledge from raw text. We propose novelty of predicted triples with respect to the training set as an important factor in interpreting results. We critically analyse the difficulty of mining novel commonsense knowledge, and show that a simple baseline method that outperforms the previous state of the art on predicting more novel triples.
Commonsense knowledge
Baseline (sea)
Base (topology)
Cite
Citations (28)
I situate the controversial critical strategies of distant reading and surface reading in the reception history of Gertrude Stein, an author whose work was frequently declared “unreadable.” I argue that an early twentieth-century history of compromised forms of reading, including women’s reading and information work, subtends both the technology with which distant reading may be carried out and the ways in which an author’s work comes to be understood as a corpus in the first place.
Cite
Citations (41)
In contrast to conventional database systems, AI systems require a knowledge base with diverse kinds of knowledge. These include, but are not limited to, knowledge about objects, knowledge about processes, and hard-to-represent commonsense knowledge about goals, motivation, causality, time, actions, etc. Attempts to represent this breadth of knowledge raise many questions: (1) How do we structure the explicit knowledge in a knowledge base? (2) How do we encode rules for manipulating a knowledge base's explicit knowledge to infer knowledge contained implicitly within the knowledge base? (3) When do we undertake and how do we control such inferences? (4) How do we formally specify the semantics of a knowledge base? (5) How do we deal with incomplete knowledge? (6) How do we extract the knowledge of an expert to initially "stock" the knowledge base? (7) How do we automatically acquire new knowledge as time goes on so that the knowledge base can be kept current? This special issue introduces this important area of artificial intelligence to a wider audience. The core of the 15 articles, contributed by a broad spectrum of researchers on various aspects of knowledge representation, show the importance, diversity, and vigor of knowledge representation as a research activity. This introduction provides some background and context to these articles by mapping out the basic approaches to knowledge representation that have developed over the years.
Commonsense knowledge
Descriptive knowledge
Cite
Citations (40)