Renmin University of China

About: Renmin University of China is a education organization based out in Beijing, Beijing, China. It is known for research contribution in the topics: China & Population. The organization has 11325 authors who have published 15498 publications receiving 238419 citations. The organization is also known as: Renmin University & People's University of China.

Recursive Visual Attention in Visual Dialog

Abstract: Visual dialog is a challenging vision-language task, which requires the agent to answer multi-round questions about an image. It typically needs to address two major problems: (1) How to answer visually-grounded questions, which is the core challenge in visual question answering (VQA); (2) How to infer the co-reference between questions and the dialog history. An example of visual co-reference is: pronouns (e.g., “they”) in the question (e.g., “Are they on or off?”) are linked with nouns (e.g., “lamps”) appearing in the dialog history (e.g., “How many lamps are there?”) and the object grounded in the image. In this work, to resolve the visual co-reference for visual dialog, we propose a novel attention mechanism called Recursive Visual Attention (RvA). Specifically, our dialog agent browses the dialog history until the agent has sufficient confidence in the visual co-reference resolution, and refines the visual attention recursively. The quantitative and qualitative experimental results on the large-scale VisDial v0.9 and v1.0 datasets demonstrate that the proposed RvA not only outperforms the state-of-the-art methods, but also achieves reasonable recursion and interpretable attention maps without additional annotations. The code is available at https://github.com/yuleiniu/rva.

Strong and Rapidly Self-Healing Hydrogels: Potential Hemostatic Materials

Abstract: Benzaldehyde-terminated telechelic four-armed polyethylene glycol (PEG-BA) is synthesized and cross-linked with carboxymethyl chitosan (CMC) to form dynamic hydrogels with strong mechanical performance. The gelation temperature and time, mechanical performance, and self-healing behaviors are systematically investigated. The hydrogels have good storage modulus up to 3162.06 ± 21.06 Pa, comparable to conventional bulk hydrogels. The separated alternate hydrogel lines connect together to become an integrated hydrogel film after 5 min at room temperature without any external intervention. This is due to the dynamic equilibrium between the Schiff base linkages and the aldehyde groups of PEG-BA and amine groups on CMC backbone. The hydrogel shows excellent cytocompatibility and the cell viability is as high as 90.7 ± 6.8% after 2 d 3D encapsulation in the hydrogel. In vivo tests indicate that the hydrogels can effectively stop bleeding when the hydrogel is directly injected into a rabbit liver incision. The total blood loss is reduced from 0.65 ± 0.10 g to 0.29 ± 0.11 g, and the hemostasis time is decreased from 167 ± 21 s to 120 ± 10 s, when compared to a gauze treatment with physical compression. These self-healing hydrogels have potential to be used as a novel hemostatic material.

Stochastic equations of non-negative processes with jumps

Abstract: We study stochastic equations of non-negative processes with jumps. The existence and uniqueness of strong solutions are established under Lipschitz and non-Lipschitz conditions. The comparison property of two solutions are proved under suitable conditions. The results are applied to stochastic equations driven by one-sided Levy processes and those of continuous state branching processes with immigration.

COCO-CN for Cross-Lingual Image Tagging, Captioning, and Retrieval

Abstract: This paper contributes to cross-lingual image annotation and retrieval in terms of data and baseline methods. We propose COCO-CN , a novel dataset enriching MS-COCO with manually written Chinese sentences and tags. For effective annotation acquisition, we develop a recommendation-assisted collective annotation system, automatically providing an annotator with several tags and sentences deemed to be relevant with respect to the pictorial content. Having 20 342 images annotated with 27 218 Chinese sentences and 70 993 tags, COCO-CN is currently the largest Chinese–English dataset that provides a unified and challenging platform for cross-lingual image tagging, captioning, and retrieval. We develop conceptually simple yet effective methods per task for learning from cross-lingual resources. Extensive experiments on the three tasks justify the viability of the proposed dataset and methods. Data and code are publicly available at https://github.com/li-xirong/coco-cn .