Learning visual representations of medical images is core to medical image understanding but its progress has been held back by the small size of hand-labeled datasets. Existing work commonly relies on transferring weights from ImageNet pretraining, which is suboptimal due to drastically different image characteristics, or rule-based label extraction from the textual report data paired with medical images, which is inaccurate and hard to generalize. We propose an alternative unsupervised strategy to learn medical visual representations directly from the naturally occurring pairing of images and textual data. Our method of pretraining medical image encoders with the paired text data via a bidirectional contrastive objective between the two modalities is domain-agnostic, and requires no additional expert input. We test our method by transferring our pretrained weights to 4 medical image classification tasks and 2 zero-shot retrieval tasks, and show that our method leads to image representations that considerably outperform strong baselines in most settings. Notably, in all 4 classification tasks, our method requires only 10% as much labeled training data as an ImageNet initialized counterpart to achieve better or comparable performance, demonstrating superior data efficiency.

Previous works related to automatic personality recognition focus on using traditional classification models with linguistic features. However, attentive neural networks with contextual embeddings, which have achieved huge success in text classification, are rarely explored for this task. In this project, we have two major contributions. First, we create the first dialogue-based personality dataset, FriendsPersona , by annotating 5 personality traits of speakers from Friends TV Show through crowdsourcing. Second, we present a novel approach to automatic personality recognition using pre-trained contextual embeddings (BERT and RoBERTa) and attentive neural networks. Our models largely improve the state-of-art results on the monologue Essays dataset by 2.49%, and establish a solid benchmark on our FriendsPersona. By comparing results in two datasets, we demonstrate the challenges of modeling personality in multi-party dialogue.

No abstract

Understanding the pore networks of unconventional tight reservoirs such as tight sandstones and shales is crucial for extracting oil/gas from such reservoirs. Mercury injection capillary pressure (MICP) and N2 gas adsorption (N2GA) are performed to evaluate pore structure of Chang-7 tight sandstone. Thin section observation, scanning electron microscope, grain size analysis, mineral composition analysis, and porosity measurement are applied to investigate geological control factors of pore structure. Grain size is positively correlated with detrital mineral content and grain size standard deviation while negatively related to clay content. Detrital mineral content and grain size are positively correlated with porosity, pore throat radius and withdrawal efficiency and negatively related to capillary pressure and pore-to-throat size ratio; while interstitial material is negatively correlated with above mentioned factors. Well sorted sediments with high debris usually possess strong compaction resistance to preserve original pores. Although many inter-crystalline pores are produced in clay minerals, this type of pores is not the most important contributor to porosity. Besides this, pore shape determined by N2GA hysteresis loop is consistent with SEM observation on clay inter-crystalline pores while BJH pore volume is positively related with clay content, suggesting N2GA is suitable for describing clay inter-crystalline pores in tight sandstones.