Search

This work studies the use of visual semantic representations to align entities in heterogeneous knowledge graphs (KGs). Images are natural components of many existing KGs. By combining visual knowledge with other auxiliary information, we show that the proposed new approach, EVA, creates a holistic entity representation that provides strong signals for cross-graph entity alignment. Besides, previous entity alignment methods require human labelled seed alignment, restricting availability. EVA provides a completely unsupervised solution by leveraging the visual similarity of entities to create an initial seed dictionary (visual pivots). Experiments on benchmark data sets DBP15k and DWY15k show that EVA offers state-of-the-art performance on both monolingual and cross-lingual entity alignment tasks. Furthermore, we discover that images are particularly useful to align long-tail KG entities, which inherently lack the structural contexts necessary for capturing the correspondences., Fangyu Liu is supported by Grace & Thomas C.H. Chan Cambridge Scholarship. This research is supported in part by the Office of the Director of National Intelligence (ODNI), Intelligence Advanced Research Projects Activity (IARPA), via IARPA Contract No. 2019- 19051600006 under the BETTER Program, and by Contracts HR0011-18-2-0052 and FA8750-19-2-1004 with the US Defence Advanced Research Projects Agency (DARPA).

Cross-target generalization constitutes an important issue for news Stance Detection (SD). In this short paper, we investigate adversarial cross-genre SD, where knowledge from annotated user-generated data is leveraged to improve news SD on targets unseen during training. We implement a BERT-based adversarial network and show experimental performance improvements over a set of strong baselines. Given the abundance of user-generated data, which are considerably less expensive to retrieve and annotate than news articles, this constitutes a promising research direction.

The Community Land Model (CLM) is the land component of the Community Earth System Model (CESM) and is used in several global and regional modeling systems. In this paper, we introduce model developments included in CLM version 5 (CLM5), which is the default land component for CESM2. We assess an ensemble of simulations, including prescribed and prognostic vegetation state, multiple forcing data sets, and CLM4, CLM4.5, and CLM5, against a range of metrics including from the International Land Model Benchmarking (ILAMBv2) package. CLM5 includes new and updated processes and parameterizations: (1) dynamic land units, (2) updated parameterizations and structure for hydrology and snow (spatially explicit soil depth, dry surface layer, revised groundwater scheme, revised canopy interception and canopy snow processes, updated fresh snow density, simple firn model, and Model for Scale Adaptive River Transport), (3) plant hydraulics and hydraulic redistribution, (4) revised nitrogen cycling (flexible leaf stoichiometry, leaf N optimization for photosynthesis, and carbon costs for plant nitrogen uptake), (5) global crop model with six crop types and time‐evolving irrigated areas and fertilization rates, (6) updated urban building energy, (7) carbon isotopes, and (8) updated stomatal physiology. New optional features include demographically structured dynamic vegetation model (Functionally Assembled Terrestrial Ecosystem Simulator), ozone damage to plants, and fire trace gas emissions coupling to the atmosphere. Conclusive establishment of improvement or degradation of individual variables or metrics is challenged by forcing uncertainty, parametric uncertainty, and model structural complexity, but the multivariate metrics presented here suggest a general broad improvement from CLM4 to CLM5.

BackgroundTopical photodynamic therapy (PDT) is widely used to treat superficial nonmelanoma skin cancer and dysplasia, and is generally well tolerated. However, as with all treatments, adverse effects may occur and awareness may facilitate approaches to prevention and management.ObjectivesTo review the available evidence relating to the adverse effects of topical PDT, to help inform recommendations in updated clinical guidelines produced by the British Association of Dermatologists and British Photodermatology Group, and the efficacy of preventative and therapeutic approaches.MethodsThis review summarizes the published evidence related to the adverse effects of topical PDT and attempts to interpret this evidence in the context of patient risk and management.ResultsPain and discomfort during PDT are acute adverse effects, which can be minimized through the use of modified and low‐irradiance PDT regimens and do not therefore usually limit successful treatment delivery. Other adverse effects include the risk of contact allergy to photosensitizer prodrugs, although this is rare but should be kept in mind, particularly for patients who have received multiple PDT treatments to larger areas. There are no other significant documented longer‐term risks and, to date, no evidence of cumulative toxicity or photocarcinogenic risk.ConclusionsTopical PDT is usually well tolerated, reinforcing the utility of this important therapeutic option in dermatology practice. The main acute adverse effect of pain can typically be minimized through preventative approaches of modified PDT regimens. Other adverse effects are uncommon and generally do not limit treatment delivery.

The 2017 Fake News Challenge Stage 1, a shared task for stance detection of news articles and claims pairs, has received a lot of attention in recent years [3]. The provided dataset is highly unbalanced, with a skewed distribution towards unrelated samples - that is, randomly generated pairs of news and claims belonging to different topics. This imbalance favored systems which performed particularly well in classifying those noisy samples, something which does not require a deep semantic understanding. In this paper, we propose a simple architecture based on conditional encoding, carefully designed to model the internal structure of a news article and its relations with a claim. We demonstrate that our model, which only leverages information from word embeddings, can outperform a system based on a large number of hand-engineered features, which replicates one of the winning systems at the Fake News Challenge [6], in the stance detection of the related samples.

Like other environmental constituents such as pressure, heat and oxygen, cold water can be either good or bad, threat or treatment depending on circumstance. Given the current increase in the popularly of open cold water swimming it is timely to review the various human responses to cold water immersion (CWI) and consider the strength of the claims made for the effects of CWI. As a consequence, in this review we look at the history of CWI and examine CWI as a pre-cursor to drowning, cardiac arrest and hypothermia. We also assess its role in prolonged survival underwater, extending exercise time in the heat and treating hyperthermic casualties. More recent uses, such as in the prevention of inflammation and treatment of inflammation-related conditions are also considered. It is concluded that the evidence-base for the different claims made for CWI are varied, and whilst in most cases there seems to be a credible rationale for the benefits or otherwise of CWI, in some cases the supporting data remain at the level of anecdotal speculation. Clear directions and requirements for future research are indicated by this review.

We put forward an approach that exploits the knowledge encoded in lexical resources in order to induce representations for words that were not encountered frequently during training. Our approach provides an advantage over the past work in that it enables vocabulary expansion not only for morphological variations, but also for infrequent domain specific terms. We performed evaluations in different settings, showing that the technique can provide consistent improvements on multiple benchmarks across domains.

Natural language processing (NLP) and named entity recognition (NER) techniques are applied to collections of newspaper articles from four cities in the U.S. Southwest. The results are used to generate a network of water management institutions that reflect public perceptions of water management and the structure of water management in these areas. This structure can be highly centralized or fragmented; in the latter case, multiple peer institutions exist that may cooperate or be in conflict. This is reflected in the public discourse of the water consumers in these areas and can, we contend, impact the potential responses of management agencies to challenges of water supply and quality and, in some cases, limit their effectiveness. Flagstaff, AZ, Tucson, AZ, Las Vegas, NV, and the Grand Valley, CO, are examined, including more than 110,000 articles from 2004–2012. Documents are scored by association with water topics, and phrases likely to be institutions are extracted via custom NLP and NER algorithms; those institutions associated with water-related documents are used to form networks via document co-location. The Grand Valley is shown to have a markedly different structure, which we contend reflects the different historical trajectory of its development and its current state, which includes multiple institutions of roughly equal scope and size. These results demonstrate the utility of using NLP and NER methods to understanding the structure and variation of water management systems.