MYMo Yu

Mo Yu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Mo Yu has a total of 12 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    A Survey of Machine Narrative Reading Comprehension Assessments

    by Yisi Sang, Xiangyang Mou, Jing Li, Jeffrey Stanton, Mo Yu

    As the body of research on machine narrative comprehension grows, there is a critical need for consideration of performance assessment strategies as well as the depth and scope of different benchmark tasks. Based on narrative theories, reading comprehension theories, as well as existing machine narrative reading comprehension tasks and datasets, we propose a typology that captures the main similarities and differences among assessment tasks; and discuss the implications of our typology for new task design and the challenges of narrative reading comprehension.


  • Open Access

    The Art of Prompting: Event Detection based on Type Specific Prompts

    by Sijia Wang, Mo Yu, Lifu Huang

    We compare various forms of prompts to represent event types and develop a unified framework to incorporate the event type specific prompts for supervised, few-shot, and zero-shot event detection. The experimental results demonstrate that a well-defined and comprehensive event type prompt can significantly improve the performance of event detection, especially when the annotated data is scarce (few-shot event detection) or not available (zero-shot event detection). By leveraging the semantics of event types, our unified framework shows up to 24.3\% F-score gain over the previous state-of-the-art baselines.

  • Open Access

    Efficient Long Sequence Encoding via Synchronization

    by Xiangyang Mou, Mo Yu, Bingsheng Yao, Lifu Huang

    Pre-trained Transformer models have achieved successes in a wide range of NLP tasks, but are inefficient when dealing with long input sequences. Existing studies try to overcome this challenge via segmenting the long sequence followed by hierarchical encoding or post-hoc aggregation. We propose a synchronization mechanism for hierarchical encoding. Our approach first identifies anchor tokens across segments and groups them by their roles in the original input sequence. Then inside Transformer layer, anchor embeddings are synchronized within their group via a self-attention module. Our approach is a general framework with sufficient flexibility -- when adapted to a new task, it is easy to be enhanced with the task-specific anchor definitions. Experiments on two representative tasks with different types of long input texts, NarrativeQA summary setting and wild multi-hop reasoning from HotpotQA, demonstrate that our approach is able to improve the global information exchange among segments while maintaining efficiency.

  • Dense Dual-Attention Network for Light Field Image Super-Resolution

    by Yu Mo, Yingqian Wang, Chao Xiao, Jungang Yang, Wei An

    Light field (LF) images can be used to improve the performance of image super-resolution (SR) because both angular and spatial information is available. It is challenging to incorporate distinctive information from different views for LF image SR. Moreover, the long-term information from the previous layers can be weakened as the depth of network increases. In this paper, we propose a dense dual-attention network for LF image SR. Specifically, we design a view attention module to adaptively capture discriminative features across different views and a channel attention module to selectively focus on informative information across all channels. These two modules are fed to two branches and stacked separately in a chain structure for adaptive fusion of hierarchical features and distillation of valid information. Meanwhile, a dense connection is used to fully exploit multi-level information. Extensive experiments demonstrate that our dense dual-attention mechanism can capture informative information across views and channels to improve SR performance. Comparative results show the advantage of our method over state-of-the-art methods on public datasets.


  • Open Access

    Toward Real-World Light Field Depth Estimation: A Noise-Aware Paradigm Using Multi-Stereo Disparity Integration

    by Yu Mo, Jungang Yang, Chao Xiao, Wei An

    Producing depth maps of a scene is essential in many light field applications. Numerous algorithms have been developed to obtain an accurate depth map on public datasets. However, real-world light field images obtained by cameras have noises produced by hardware limitations. Existing methods do not perform well on noisy scenes. In this paper, we propose a noise-aware light field depth estimation algorithm which is insensitive to noise. First, an optical flow approach is used to compute disparities between the central view and each of the remaining views, then a weighted integration method is used to reduce calculation errors. Second, a refinement framework, which is measured using the refocus image and angular patch, is proposed to handle occlusion and texture-less parts. Third, a K-means clustering-based strategy is used to improve noise capability. Finally, a depth map is regularized by a Markov random field (MRF). Extensive experiments are conducted on synthetic datasets and real-world datasets. Both quantitative and qualitative results have demonstrated the superiority and robustness of our method.