Explore the vast range of titles available.

Heavy precipitation events and fluvial flood can cause serious damage to property and fatality, and thus flood disaster management is essential to lower the risk, to prevent and to mitigate the crisis. Effective and efficient flood disaster management requires on-site information, forecast and appropriate corresponding response strategies and measures. Therefore, a concept of integrated decision support system (DSS) for flood disaster management is proposed and discussed in this paper. The DSS integrates observation, rainfall forecast, fast flood simulation, historical events and crisis scenarios, and response strategies and measures, by means of information and communication technology (ICT) and internet of things (IoT). The DSS can provide relevant information and offer suggestions of action before and during the crisis. The DSS can also be utilised for disaster drill, education, and demonstration with the help of virtual reality (VR), augmented reality (AR) or Metaverse technology. The aims of this DSS are to improve existing response strategies and emergency services, to enhance the community-based disaster risk management, and to raise public crisis awareness.

Ozone micron bubble (OMB) treatment offers a promising approach to effectively eliminate Antibiotic Resistance Genes (ARGs) from infectious medical wastewater and mitigate the threat of drug resistance transmission. This study evaluated the effectiveness of OMB treatment for reducing ARGs from infectious medical wastewater in laboratory and on-site pilot treatment setups. In part, the presence of antibiotic residues in a hospital wastewater treatment plant (WWTP) and the impact of hospital wastewater on the distribution of ARGs in a wastewater collection system were also investigated. The results of wastewater collection system survey revealed a high prevalence of ARGs in the system, particularly mcr- 1, largely originating from medical wastewater discharges. Furthermore, analysis of antibiotic residues in the hospital wastewater treatment system showed significant accumulation, particularly of quinolone antibiotics, in the biomass of the biological oxidation tank, suggesting a potential risk of ARG proliferation within the system. Comparison of wastewater samples from domestic and hospital WWTPs revealed a relatively higher abundance of ARGs in the latter, with differences ranging from 2.2 to sixfold between corresponding locations in the treatment plants. Notably, the biological oxidation unit of both WWTPs exhibited a greater proportion of ARGs among all sampled points, indicating the potential proliferation of ARGs within the biomass of the treatment units. ARG degradation experiments showed that OMB treatment resulted in a significantly lower CT value (9.3 mg O 3 L −1  min) compared to ozone coarse bubble treatment (102 mg O 3 L −1  min) under identical test conditions. Moreover, the use of OMB on site significantly reduced the accumulation of ARGs in hospital wastewater, underscoring its potential as an effective solution for mitigating ARG spread.

In recent advancements in spoken question answering (QA), end-to-end models have made significant strides. However, previous research has primarily focused on extractive span selection. While this extractive-based approach is effective when answers are present directly within the input, it falls short in addressing abstractive questions, where answers are not directly extracted but inferred from the given information. To bridge this gap, we introduce the first end-to-end Generative Spoken Question Answering (GSQA) model that empowers the system to engage in abstractive reasoning. The challenge in training our GSQA model lies in the absence of a spoken abstractive QA dataset. We propose using text models for initialization and leveraging the extractive QA dataset to transfer knowledge from the text generative model to the spoken generative model. Experimental results indicate that our model surpasses the previous extractive model by 3% on extractive QA datasets. Furthermore, the GSQA model has only been fine-tuned on the spoken extractive QA dataset. Despite not having seen any spoken abstractive QA data, it can still closely match the performance of the cascade model. In conclusion, our GSQA model shows the potential to generalize to a broad spectrum of questions, thus further expanding the spoken question answering capabilities of abstractive QA. Our code is available at https://voidful.github.io/GSQA

This paper presents Meta-Whisper, a novel approach to improve automatic speech recognition (ASR) for low-resource languages using the Whisper model. By leveraging Meta In-Context Learning (Meta-ICL) and a k-Nearest Neighbors (KNN) algorithm for sample selection, Meta-Whisper enhances Whisper's ability to recognize speech in unfamiliar languages without extensive fine-tuning. Experiments on the ML-SUPERB dataset show that Meta-Whisper significantly reduces the Character Error Rate (CER) for low-resource languages compared to the original Whisper model. This method offers a promising solution for developing more adaptable multilingual ASR systems, particularly for languages with limited resources.

Ever since the development of GPT-3 in the natural language processing (NLP) field, in-context learning (ICL) has played an essential role in utilizing large language models (LLMs). By presenting the LM utterance-label demonstrations at the input, the LM can accomplish few-shot learning without relying on gradient descent or requiring explicit modification of its parameters. This enables the LM to perform various downstream tasks in a black-box manner. Despite the success of ICL in NLP, little work is exploring the possibility of ICL in speech processing. This study is the first work exploring ICL for speech classification tasks with textless speech LM. We first show that the current speech LM lacks the ICL capability. We then perform warmup training on the speech LM, equipping the LM with demonstration learning capability. This paper explores and proposes the first speech LM capable of performing unseen classification tasks in an ICL manner.

Imitation learning addresses the challenge of learning by observing an expert's demonstrations without access to reward signals from environments. Most existing imitation learning methods that do not require interacting with environments either model the expert distribution as the conditional probability p(a|s) (e.g., behavioral cloning, BC) or the joint probability p(s, a). Despite the simplicity of modeling the conditional probability with BC, it usually struggles with generalization. While modeling the joint probability can improve generalization performance, the inference procedure is often time-consuming, and the model can suffer from manifold overfitting. This work proposes an imitation learning framework that benefits from modeling both the conditional and joint probability of the expert distribution. Our proposed Diffusion Model-Augmented Behavioral Cloning (DBC) employs a diffusion model trained to model expert behaviors and learns a policy to optimize both the BC loss (conditional) and our proposed diffusion model loss (joint). DBC outperforms baselines in various continuous control tasks in navigation, robot arm manipulation, dexterous manipulation, and locomotion. We design additional experiments to verify the limitations of modeling either the conditional probability or the joint probability of the expert distribution, as well as compare different generative models. Ablation studies justify the effectiveness of our design choices.

Prior work has demonstrated that data augmentation is useful for improving dialogue state tracking. However, there are many types of user utterances, while the prior method only considered the simplest one for augmentation, raising the concern about poor generalization capability. In order to better cover diverse dialogue acts and control the generation quality, this paper proposes controllable user dialogue act augmentation (CUDA-DST) to augment user utterances with diverse behaviors. With the augmented data, different state trackers gain improvement and show better robustness, achieving the state-of-the-art performance on MultiWOZ 2.1

Currently, the most active electrocatalysts for the conversion of CO₂ to CO are gold-based nanomaterials, whereas non–precious metal catalysts have shown low to modest activity. Here, we report a catalyst of dispersed single-atom iron sites that produces CO at an overpotential as low as 80 millivolts. Partial current density reaches 94 milliamperes per square centimeter at an overpotential of 340 millivolts. Operando x-ray absorption spectroscopy revealed the active sites to be discrete Fe3+ ions, coordinated to pyrrolic nitrogen (N) atoms of the N-doped carbon support, that maintain their +3 oxidation state during electrocatalysis, probably through electronic coupling to the conductive carbon support. Electrochemical data suggest that the Fe3+ sites derive their superior activity from faster CO₂ adsorption and weaker CO absorption than that of conventional Fe2+ sites.

Unraveling the precise location and nature of active sites is of paramount significance for the understanding of the catalytic mechanism and the rational design of efficient electrocatalysts. Here, we use well-defined crystalline cobalt oxyhydroxides CoOOH nanorods and nanosheets as model catalysts to investigate the geometric catalytic active sites. The morphology-dependent analysis reveals a ~50 times higher specific activity of CoOOH nanorods than that of CoOOH nanosheets. Furthermore, we disclose a linear correlation of catalytic activities with their lateral surface areas, suggesting that the active sites are exclusively located at lateral facets rather than basal facets. Theoretical calculations show that the coordinatively unsaturated cobalt sites of lateral facets upshift the O 2p -band center closer to the Fermi level, thereby enhancing the covalency of Co-O bonds to yield the reactivity. This work elucidates the geometrical catalytic active sites and enlightens the design strategy of surface engineering for efficient OER catalysts. While cobalt-based electrocatalysts demonstrate promising performances for oxygen evolution, active site identification is complicated by concurrent structural changes. Here, authors examine crystalline, well-defined cobalt oxyhydroxide nanomaterials and identify the geometric active sites.

Van der Waals heterobilayers of transition metal dichalcogenides with spin–valley coupling of carriers in different layers have emerged as a new platform for exploring spin/valleytronic applications. The interlayer coupling was predicted to exhibit subtle changes with the interlayer atomic registry. Manually stacked heterobilayers, however, are incommensurate with the inevitable interlayer twist and/or lattice mismatch, where the properties associated with atomic registry are difficult to access by optical means. Here, we unveil the distinct polarization properties of valley-specific interlayer excitons using epitaxially grown, commensurate WSe 2 /MoSe 2 heterobilayers with well-defined (AA and AB) atomic registry. We observe circularly polarized photoluminescence from interlayer excitons, but with a helicity opposite to the optical excitation. The negative circular polarization arises from the quantum interference imposed by interlayer atomic registry, giving rise to distinct polarization selection rules for interlayer excitons. Using selective excitation schemes, we demonstrate the optical addressability for interlayer excitons with different valley configurations and polarization helicities. The interlayer coupling in van der Waals heterostructures is sensitive to the interlayer atomic registry. Here, the authors investigate the polarisation properties of epitaxially grown, commensurate WSe 2 /MoSe 2 heterobilayers with well-defined atomic registry, and observe negative, circularly polarized photoluminescence from interlayer excitons.