A5100683043- Chinese history and philosophy
- Japanese History and Culture
- Digital Marketing and Social Media
- Human Mobility and Location-Based Analysis
- Optical Network Technologies
- China's Socioeconomic Reforms and Governance
- Complex Network Analysis Techniques
- Sentiment Analysis and Opinion Mining
- Advanced Surface Polishing Techniques
- Advanced Measurement and Metrology Techniques
- Natural Language Processing Techniques
- Neural Networks and Reservoir Computing
- Laser Material Processing Techniques
- Speech and dialogue systems
- Photonic and Optical Devices
Fudan University
2016-2020
University of Science and Technology of China
2019
Predicting the popularity of online article sheds light to many applications such as recommendation, advertising and information retrieval. However, there are several technical challenges be addressed for developing best predictive capability. (1) The fluctuates under impacts external factors, which unpredictable hard capture. (2) Content meta-data features, largely determining content popularity, usually multi-modal nontrivial model. (3) Besides, it also needs figure out how integrate...
Recently, there has been tremendous research studies in optical neural networks that could complete comparatively complex computation by characteristic with much more fewer dissipation than electrical networks. Existed based on the circuit are structured an grating platform different diffractive phases at points (Chen and Zhu, 2019 Mo et al., 2018). In this study, it proposed a multiwave deep network approximately 10 6 synapses, is easy to make hardware implementation of neuromorphic...
Evaluating instruction following capabilities for multimodal, multi-turn dialogue is challenging. With potentially multiple instructions in the input model context, task time-consuming human raters and we show LLM based judges are biased towards answers from same model. We propose MMMT-IF, an image Q$\&$A evaluation set with added global between questions, constraining answer format. This challenges models to retrieve dispersed across long dialogues reason under constraints. All objectively...