Explore the vast range of titles available.

Retrieval Augment Generation (RAG) is a recent advancement in Open-Domain Question Answering (ODQA). RAG has only been trained and explored with a Wikipedia-based external knowledge base and is not optimized for use in other specialized domains such as healthcare and news. In this paper, we evaluate the impact of joint training of the retriever and generator components of RAG for the task of domain adaptation in ODQA. We propose , an extension to RAG that can adapt to a domain-specific knowledge base by updating all components of the external knowledge base during training. In addition, we introduce an auxiliary training signal to inject more domain-specific knowledge. This auxiliary signal forces to reconstruct a given sentence by accessing the relevant information from the external knowledge base. Our novel contribution is that, unlike RAG, RAG-end2end does joint training of the retriever and generator for the end QA task and domain adaptation. We evaluate our approach with datasets from three domains: COVID-19, News, and Conversations, and achieve significant performance improvements compared to the original RAG model. Our work has been open-sourced through the HuggingFace Transformers library, attesting to our work’s credibility and technical consistency.

This paper examines how existing knowledge base (i.e., knowledge breadth and depth) interacts with knowledge integration mechanisms (i.e., external market knowledge acquisition and internal knowledge sharing) to affect radical innovation. Survey data from high technology companies in China demonstrate that the effects of knowledge breadth and depth are contingent on market knowledge acquisition and knowledge sharing in opposite ways. In particular, a firm with a broad knowledge base is more likely to achieve radical innovation in the presence of internal knowledge sharing rather than market knowledge acquisition. In contrast, a firm with a deep knowledge base is more capable of developing radical innovation through market knowledge acquisition rather than internal knowledge sharing.

Purpose Enhancing green innovation for corporate sustainability is one of the recent issues globally. Knowledge management has been determined as a core factor that hamstrings green innovation. The existing literature was limited to expose the importance of the knowledge management process for corporate sustainable performance. Thus, this paper aims to examine the role of the knowledge management process for corporate sustainable performance with the integration of green innovation and organizational agility following the resource-based view theory. Design/methodology/approach Cross-sectional design was used in this study. Data were gathered through convenience sampling from 475 respondents of multinational manufacturing corporations of Pakistan, analyzed by using structural equation modeling. Findings This study revealed that the knowledge management process and its all constructs (acquisition, dissemination and application) lead toward green innovation; further, green innovation influences corporate sustainable performance and its all constructs (environment, economic and social). Green innovation partially mediates the association between the knowledge management process and corporate sustainable performance. Besides, organizational agility has a positive effect on green innovation and corporate sustainable performance but was not found moderating these relations. The study educates that organizations investing in innovative technologies and adopting greener strategies are not only adequate for achieving sustainable performance, soft issues such as knowledge management and organizational agility but also important factors in the current knowledge base economy. Originality/value This study is an attempt to examine the previously undiscovered multi-dimensional relationships among the knowledge management process, green innovation, organizational agility and corporate sustainable performance. The presence of a positive correlation among these constructs was observed, proving the conceptual framework for this study.

Purpose This paper aims to study the effects of knowledge acquisition on innovation performance and the moderating effects of human resource management (HRM), in terms of employee retention and HRM practices, on the above-mentioned relationship. Design/methodology/approach A sample of 129 firms operating in a wide array of sectors has been used to gather data through a standardized questionnaire for testing the hypotheses through ordinary least squares (OLS) regression models. Findings The results indicate that knowledge acquisition positively affects innovation performance and that HRM moderates the relationship between knowledge acquisition and innovation performance. Originality/value With the increasing proclivity towards engaging in open innovation, firms are likely to face some tensions and opportunities leading to a shift in the management of human resources. This starts from the assumption that the knowledge base of the firm resides in the people who work for the firm and that some HRM factors can influence innovation within firms. Despite this, there is a lack of research investigating the link between knowledge acquisition, HRM and innovation performance under the open innovation lens. This paper intends to fill this gap and nurture future research by assessing whether knowledge acquisition influences innovation performance and whether HRM moderates such a relationship.

Providing model-generated explanations in recommender systems is important to user experience. State-of-the-art recommendation algorithms—especially the collaborative filtering (CF)- based approaches with shallow or deep models—usually work with various unstructured information sources for recommendation, such as textual reviews, visual images, and various implicit or explicit feedbacks. Though structured knowledge bases were considered in content-based approaches, they have been largely ignored recently due to the availability of vast amounts of data and the learning power of many complex models. However, structured knowledge bases exhibit unique advantages in personalized recommendation systems. When the explicit knowledge about users and items is considered for recommendation, the system could provide highly customized recommendations based on users’ historical behaviors and the knowledge is helpful for providing informed explanations regarding the recommended items. A great challenge for using knowledge bases for recommendation is how to integrate large-scale structured and unstructured data, while taking advantage of collaborative filtering for highly accurate performance. Recent achievements in knowledge-base embedding (KBE) sheds light on this problem, which makes it possible to learn user and item representations while preserving the structure of their relationship with external knowledge for explanation. In this work, we propose to explain knowledge-base embeddings for explainable recommendation. Specifically, we propose a knowledge-base representation learning framework to embed heterogeneous entities for recommendation, and based on the embedded knowledge base, a soft matching algorithm is proposed to generate personalized explanations for the recommended items. Experimental results on real-world e-commerce datasets verified the superior recommendation performance and the explainability power of our approach compared with state-of-the-art baselines.

To the Editor: Over the past two decades, the scale and complexity of genomics technologies and data have advanced from sequencing genomes of a few organisms to generating metagenomes, genome variation, gene expression, metabolites, and phenotype data for thousands of organisms and their communities. A major challenge in this data-rich age of biology is integrating heterogeneous and distributed data into predictive models of biological function, ranging from a single gene to entire organisms and their ecologies. Here we present the DOE Systems Biology Knowledgebase (KBase, http://kbase.us), an open-source software and data platform that enables data sharing, integration, and analysis of microbes, plants, and their communities. Once a Narrative has been shared or made public, other users can copy the Narrative and rerun it on their own data, or modify it to suit their scientific needs. [...]public Narratives serve as resources for the user community by capturing valuable data sets, associated computational analyses, and scientific context describing the rationale behind a scientific study in a form that is immediately reproducible and reusable.

Local and indigenous knowledge is being transformed globally, particularly being eroded when pertaining to ecology. In many parts of the world, rural and indigenous communities are facing tremendous cultural, economic and environmental changes, which contribute to weaken their local knowledge base. In the face of profound and ongoing environmental changes, both cultural and biological diversity are likely to be severely impacted as well as local resilience capacities from this loss. In this global literature review, we analyse the drivers of various types of local and indigenous ecological knowledge transformation and assess the directionality of the reported change. Results of this analysis show a global impoverishment of local and indigenous knowledge with 77% of papers reporting the loss of knowledge driven by globalization, modernization, and market integration. The recording of this loss, however, is not symmetrical, with losses being recorded more strongly in medicinal and ethnobotanical knowledge. Persistence of knowledge (15% of the studies) occurred in studies where traditional practices were being maintained consiously and where hybrid knowledge was being produced as a resut of certain types of incentives created by economic development. This review provides some insights into local and indigenous ecological knowledge change, its causes and implications, and recommends venues for the development of replicable and comparative research. The larger implication of these results is that because of the interconnection between cultural and biological diversity, the loss of local and indigenous knowledge is likely to critically threaten effective conservation of biodiversity, particularly in community-based conservation local efforts.

Background The International Classification of Diseases (ICD) has long been the main basis for comparability of statistics on causes of mortality and morbidity between places and over time. This paper provides an overview of the recently completed 11th revision of the ICD, focusing on the main innovations and their implications. Main text Changes in content reflect knowledge and perspectives on diseases and their causes that have emerged since ICD-10 was developed about 30 years ago. Changes in design and structure reflect the arrival of the networked digital era, for which ICD-11 has been prepared. ICD-11’s information framework comprises a semantic knowledge base (the Foundation), a biomedical ontology linked to the Foundation and classifications derived from the Foundation. ICD-11 for Mortality and Morbidity Statistics (ICD-11-MMS) is the primary derived classification and the main successor to ICD-10. Innovations enabled by the new architecture include an online coding tool (replacing the index and providing additional functions), an application program interface to enable remote access to ICD-11 content and services, enhanced capability to capture and combine clinically relevant characteristics of cases and integrated support for multiple languages. Conclusions ICD-11 was adopted by the World Health Assembly in May 2019. Transition to implementation is in progress. ICD-11 can be accessed at icd.who.int.

We present mGENRE, a sequence-to- sequence system for the Multilingual Entity Linking (MEL) problem—the task of resolving language-specific mentions to a multilingual Knowledge Base (KB). For a mention in a given language, mGENRE predicts the name of the target entity left-to-right, token-by-token in an autoregressive fashion. The autoregressive formulation allows us to effectively cross-encode mention string and entity names to capture more interactions than the standard dot product between mention and entity vectors. It also enables fast search within a large KB even for mentions that do not appear in mention tables and with no need for large-scale vector indices. While prior MEL works use a single representation for each entity, we match against entity names of as many languages as possible, which allows exploiting language connections between source input and target name. Moreover, in a zero-shot setting on languages with no training data at all, mGENRE treats the target language as a latent variable that is marginalized at prediction time. This leads to over 50% improvements in average accuracy. We show the efficacy of our approach through extensive evaluation including experiments on three popular MEL benchmarks where we establish new state-of-the-art results. Source code available at .

To develop a knowledge-aware recommender system, a key issue is how to obtain rich and structured knowledge base (KB) information for recommender system (RS) items. Existing data sets or methods either use side information from original RSs (containing very few kinds of useful information) or utilize a private KB. In this paper, we present , a data set linking KB information for RSs. It has linked three widely used RS data sets with two popular KBs, namely Freebase and YAGO. Based on our linked data set, we first preform qualitative analysis experiments, and then we discuss the effect of two important factors (i.e., popularity and recency) on whether a RS item can be linked to a KB entity. Finally, we compare several knowledge-aware recommendation algorithms on our linked data set.