A5039125805- Topic Modeling
- Natural Language Processing Techniques
- Protein Tyrosine Phosphatases
- Head and Neck Cancer Studies
- Advanced Text Analysis Techniques
- Angiogenesis and VEGF in Cancer
- ATP Synthase and ATPases Research
- Wnt/β-catenin signaling in development and cancer
- Multimodal Machine Learning Applications
- Speech and dialogue systems
- Acute Myeloid Leukemia Research
- Biodiesel Production and Applications
- Mechanisms of cancer metastasis
- Advanced Combustion Engine Technologies
- Neural Networks and Applications
- Music and Audio Processing
- Phagocytosis and Immune Regulation
- Immune cells in cancer
- Kruppel-like factors research
- PI3K/AKT/mTOR signaling in cancer
- Remote Sensing and LiDAR Applications
- RNA Interference and Gene Delivery
- Reconstructive Facial Surgery Techniques
- Nasal Surgery and Airway Studies
- AI in Service Interactions
University of Electronic Science and Technology of China
2024
First Affiliated Hospital of Jiangxi Medical College
2022-2024
Shanghai Ninth People's Hospital
2024
University of Illinois Urbana-Champaign
2020-2023
Fifth Affiliated Hospital of Sun Yat-sen University
2022-2023
Jiangxi Provincial People's Hospital
2020-2023
Sun Yat-sen University
2022-2023
Beijing Children’s Hospital
2022-2023
Capital Medical University
2022-2023
First Affiliated Hospital of Gannan Medical University
2023
Jingjing Xu, Xu Sun, Qi Zeng, Xiaodong Zhang, Xuancheng Ren, Houfeng Wang, Wenjie Li. Proceedings of the 56th Annual Meeting Association for Computational Linguistics (Volume 1: Long Papers). 2018.
Narrative story generation is a challenging problem because it demands the generated sentences with tight semantic connections, which has not been well studied by most existing generative models. To address this problem, we propose skeleton-based model to promote coherence of stories. Different from traditional models that generate complete sentence at stroke, proposed first generates critical phrases, called skeleton, and then expands skeleton fluent sentence. The manually defined, but...
We introduce a new task, MultiMedia Event Extraction, which aims to extract events and their arguments from multimedia documents. develop the first benchmark collect dataset of 245 news articles with extensively annotated arguments. propose novel method, Weakly Aligned Structured Embedding (WASE), that encodes structured representations semantic information textual visual data into common embedding space. The structures are aligned across modalities by employing weakly supervised training...
Manling Li, Qi Zeng, Ying Lin, Kyunghyun Cho, Heng Ji, Jonathan May, Nathanael Chambers, Clare Voss. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP). 2020.
Most existing works on dialog systems only consider conversation content while neglecting the personality of user bot is interacting with, which begets several unsolved issues. In this paper, we present a personalized end-to-end model in an attempt to leverage personalization goal-oriented dialogs. We first introduce PROFILE MODEL encodes profiles into distributed embeddings and refers history from other similar users. Then PREFERENCE captures preferences over knowledge base entities handle...
The goal of sentiment-to-sentiment "translation" is to change the underlying sentiment a sentence while keeping its content. main challenge lack parallel data. To solve this problem, we propose cycled reinforcement learning method that enables training on unpaired data by collaboration between neutralization module and an emotionalization module. We evaluate our approach two review datasets, Yelp Amazon. Experimental results show significantly outperforms state-of-the-art systems....
Jingjing Xu, Yuechen Wang, Duyu Tang, Nan Duan, Pengcheng Yang, Qi Zeng, Ming Zhou, Xu Sun. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint (EMNLP-IJCNLP). 2019.
Generating semantically coherent responses is still a major challenge in dialogue generation. Different from conventional text generation tasks, the mapping between inputs and conversations more complicated, which highly demands understanding of utterance-level semantic dependency, relation whole meanings outputs. To address this problem, we propose an Auto-Encoder Matching (AEM) model to learn such dependency. The contains two auto-encoders one module. representations responses, module...
This study aimed to investigate the effects of drinking water with high concentrations iodine on intelligence children in Tianjin, China.It was a population-based health survey utilizing random cluster sampling design conducted June 2005. Participants were recruited from total population primary school attending years 1-4 ages ranging 8 10 years. Intelligence quotient (IQ) assessed using combined Raven's test, second edition. Linear regression analyses applied test for any association...
Xinya Du, Zixuan Zhang, Sha Li, Pengfei Yu, Hongwei Wang, Tuan Lai, Xudong Lin, Ziqi Iris Liu, Ben Zhou, Haoyang Wen, Manling Darryl Hannan, Jie Lei, Hyounghun Kim, Rotem Dror, Haoyu Michael Regan, Qi Zeng, Qing Lyu, Charles Carl Edwards, Xiaomeng Jin, Yizhu Jiao, Ghazaleh Kazeminejad, Zhenhailong Chris Callison-Burch, Mohit Bansal, Vondrick, Jiawei Han, Dan Roth, Shih-Fu Chang, Martha Palmer, Heng Ji. Proceedings of the 2022 Conference North American Chapter Association for Computational...
Events are inter-related in documents. Motivated by the one-sense-per-discourse theory, we hypothesize that a participant tends to play consistent roles across multiple events same document. However recent work on document-level event argument extraction models each individual isolation and therefore causes inconsistency among extracted arguments events, which will further cause discrepancy for downstream applications such as knowledge base population, question answering, hypothesis...
Jingjing Xu, Liang Zhao, Hanqi Yan, Qi Zeng, Yun Liang, Xu Sun. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint (EMNLP-IJCNLP). 2019.
Abstract The two-dimensional (2D) lidar is a ranging optical sensor that can measure the cross-section of geometric structure environment. We propose robust 2D simultaneous localization and mapping (SLAM) algorithm working in ambiguous environments. To improve front-end scan-matching module’s accuracy robustness, we performing degeneration analysis, line landmark tracking, environment coverage analysis. max-clique selection odometer verification are introduced to increase stability SLAM an...
Background: The repair of nasal alar defects is challenging for plastic surgeons, and there currently no standard operation. Herein, the authors reported clinical outcomes a nasofacial groove pedicled flap reconstruction defect. Methods: This retrospective study included patients who underwent defect between January 2018 June 2020. Photographs facial postures were taken before after surgery to record surgical results patients. patient’s medical history was reviewed retrospectively....
As one of the most important types evidence at scene crime, rapid identification human bloodstain is great significance to solve criminal case. In this paper, spectral data different species samples including human, chicken and pig were acquired by using a hand-held near-infrared spectrometer. Then, training models established via convolutional neural network-support vector machine algorithm. Meanwhile, traditional support machine, genetic algorithm-back propagation random forest...
Existing factual consistency evaluation approaches for text summarization provide binary predictions and limited insights into the weakness of systems. Therefore, we propose task fine-grained inconsistency detection, goal which is to predict types errors in a summary. Motivated by how humans inspect summaries, an interpretable detection model, FineGrainFact, explicitly represents facts documents summaries with semantic frames extracted role labeling, highlights related inconsistency. The...
Guangxiang Zhao, Jingjing Xu, Qi Zeng, Xuancheng Ren, Xu Sun. Proceedings of the 2019 Conference North American Chapter Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers). 2019.
Massive thymic hyperplasia (MTH) is a very rare entity, with fewer than 20 cases reported in the literature infancy. Most patients have respiratory symptoms and enlarged thymus gland occupies one side of thoracic cavity. Posterolateral thoracotomy or median sternotomy main treatment for MTH infants. We report case an infant which occupied his bilateral cavity he underwent video-assisted thoracoscopic surgery (VATS). In addition, we reviewed summarized relevant literature.A 4-month-old boy...
Liliang Ren, Mankeerat Sidhu, Qi Zeng, Revanth Gangi Reddy, Heng Ji, ChengXiang Zhai. Proceedings of the Third DialDoc Workshop on Document-grounded Dialogue and Conversational Question Answering. 2023.
A variety of factors affect English classroom teaching, which prevents teachers from effectively grasping students’ learning status and situation. In particular, management is more difficult during online teaching. order to improve the effectiveness based on human-computer interaction algorithm facial identification algorithm, this paper recognizes process students in eliminates image background according actual teaching needs. Moreover, by extracting fusing time sequence information spatial...
Current methods for event representation ignore related events in a corpus-level global context. For deep and comprehensive understanding of complex events, we introduce new task, Event Network Embedding, which aims to represent by capturing the connections among events. We propose novel framework, Global Embedding (GENE), that encodes network with multi-view graph encoder while preserving topology node semantics. The is trained minimizing both structural semantic losses. develop series...