Fine-Grained Similarity Measurement between Educational Videos and Exercises

Wang, Xin; Huang, Wei; Liu, Qi; Yin, Yan; Huang, Zhenya; Wu, Le; Ma, Jinghong; Wang, Xue

doi:10.1145/3394171.3413783

Cited by 16 publications

(4 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Inspired by [25] [26] , we separate vectors into similar and dissimilar components based on sentence similarity learning method. To be specific, two phases are included: decomposition and composition.…”

Section: ) Semantic Decomposition Layermentioning

confidence: 99%

NeurReview: A Neural Architecture Based Conformity Prediction of Peer Reviews

Meng

2023

IEEE Access

View full text Add to dashboard Cite

Peer review is at the heart of scholarly communications and the foundation of scientific evaluation. However, peer review's effectiveness is continuously challenged due to biased and inconsistent peer reviews. Consequently, ensuring the quality of peer reviews is a time-critical problem. In this paper, we investigate the conformity between reviews and meta-reviews. To predict the review conformity and identify the effective features to distinguish the misaligned reviews, we propose NeurReview, which models the review process from the review structure and interactions with authors and other reviewers. Two evaluation datasets are constructed from the ICLR open reviews. The evaluation results verified the efficacy of our proposed model. In addition, we found that the divergence with other reviews and responses, the consistency of sentiment polarity with the recommendation score, etc., are beneficial features for identifying lowconformity reviews, which can assist meta-reviewers in making final decisions.

show abstract

Section: ) Semantic Decomposition Layermentioning

confidence: 99%

NeurReview: A Neural Architecture Based Conformity Prediction of Peer Reviews

Meng

2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Further interesting articles proposed methods to create videos automatically [65,87,88], alleviate the split attention problem [81,209], enhance speech recognition [92,202], and find content that is similar not in other videos but in other media such as in textbooks [252] and books [97].…”

Section: Othermentioning

confidence: 99%

A Closer Look into Recent Video-based Learning Research: A Comprehensive Review of Video Characteristics, Tools, Technologies, and Learning Effectiveness

Navarrete¹,

Nehring²,

Schanze³

et al. 2023

Preprint

View full text Add to dashboard Cite

People increasingly use videos on the Web as a source for learning, be it for daily tasks or in formal as well as informal educational settings. To support this way of learning, researchers and developers are continuously developing tools, proposing guidelines, analyzing data, and conducting experiments. However, it is still not clear what characteristics a video should have to be an effective learning medium. In this paper, we present a comprehensive review of 257 articles on video-based learning from computer science databases for the period from 2016 to 2021 using the PRISMA guidelines. One of the aims of the review is to identify the video characteristics that have been explored by previous work. Based on our analysis, we suggest a taxonomy which organizes the video characteristics and contextual aspects into eight categories: (1) audio features, (2) visual features, (3) textual features, (4) instructor behavior, (5) learners' activities, (6) interactive features (quizzes, etc.), (7) production style, and (8) instructional design. Also, we identify four representative research directions: (1) proposals of tools to support video-based learning, (2) studies with controlled experiments, (3) data analysis studies, and (4) proposals of design guidelines for learning videos. We find that the most explored characteristics are textual features followed by visual features, learner's activities, and interactive features. Text of transcripts, video frames, and images (figures and illustrations) are most frequently used by tools that support learning through videos. The learner's activity is heavily explored through log files in data analysis studies, and interactive features have been frequently scrutinized in controlled experiments. We complement our review by contrasting research findings that investigate the impact of video characteristics on the learning effectiveness, report on tasks and technologies used to develop tools that support learning, and summarize trends in the development of design guidelines to produce learning videos.

show abstract

“…Although the modern deep learning network (DNN) has shown good performance on the task at hand [5,12,24,41,46,47], they always violently forget the previous knowledge when they learn new tasks. Such a phenomenon is known as catastrophic forgetting, which is the research hotspot in Continual Learning (CL) [11,28,32].…”

Section: Introductionmentioning

confidence: 99%

Class Gradient Projection For Continual Learning

Chen

Zhang

Song

et al. 2022

Proceedings of the 30th ACM International Conference on Multimedia

View full text Add to dashboard Cite

Catastrophic forgetting is one of the most critical challenges in Continual Learning (CL). Recent approaches tackle this problem by projecting the gradient update orthogonal to the gradient subspace of existing tasks. While the results are remarkable, those approaches ignore the fact that these calculated gradients are not guaranteed to be orthogonal to the gradient subspace of each class due to the class deviation in tasks, e.g., distinguishing "Man" from "Sea" v.s. differentiating "Boy" from "Girl". Therefore, this strategy may still cause catastrophic forgetting for some classes. In this paper, we propose Class Gradient Projection (CGP), which calculates the gradient subspace from individual classes rather than tasks. Gradient update orthogonal to the gradient subspace of existing classes can be effectively utilized to minimize interference from other classes. To improve the generalization and efficiency, we further design a Base Refining (BR) algorithm to combine similar classes and refine class bases dynamically. Moreover, we leverage a contrastive learning method to improve the model's ability to handle unseen tasks. Extensive experiments on benchmark datasets demonstrate the effectiveness of our proposed approach. It improves the previous methods by 2.0% on the CIFAR-100 dataset. The code is available at https://github.com/zackschen/CGP. CCS CONCEPTS• Computing methodologies → Image representations.

show abstract

Fine-Grained Similarity Measurement between Educational Videos and Exercises

Cited by 16 publications

References 38 publications

NeurReview: A Neural Architecture Based Conformity Prediction of Peer Reviews

NeurReview: A Neural Architecture Based Conformity Prediction of Peer Reviews

A Closer Look into Recent Video-based Learning Research: A Comprehensive Review of Video Characteristics, Tools, Technologies, and Learning Effectiveness

Class Gradient Projection For Continual Learning

Contact Info

Product

Resources

About