Search

Retrieval-Augmented Generation (RAG) has been shown to enhance the factual accuracy of Large Language Models (LLMs), but existing methods often suffer from limited reasoning capabilities in effectively using the retrieved evidence, particularly when using open-source LLMs. To mitigate this gap, we introduce a novel framework, Open-RAG, designed to enhance reasoning capabilities in RAG with open-source LLMs. Our framework transforms an arbitrary dense LLM into a parameter-efficient sparse mixture of experts (MoE) model capable of handling complex reasoning tasks, including both single- and multi-hop queries. Open-RAG uniquely trains the model to navigate challenging distractors that appear relevant but are misleading. As a result, Open-RAG leverages latent learning, dynamically selecting relevant experts and integrating external knowledge effectively for more accurate and contextually relevant responses. In addition, we propose a hybrid adaptive retrieval method to determine retrieval necessity and balance the trade-off between performance gain and inference speed. Experimental results show that the Llama2-7B-based Open-RAG outperforms state-of-the-art LLMs and RAG models such as ChatGPT, Self-RAG, and Command R+ in various knowledge-intensive tasks. We open-source our code and models at https://openragmoe.github.io/, Comment: Accepted to EMNLP 2024 Findings. Website: https://openragmoe.github.io/. 14 pages, 7 figures, 5 tables

Cervical cancer affects millions of women worldwide and has a significantly higher survival rate when diagnosed early. Pap smears and cervical biopsies are vital screening tools for detecting such cancer. However, the success of these screening processes depends on the skills of cytologists. A recent trend in diagnostic cytology is to apply machine-learning-based models to classify cancer using cell images. These automated models have been shown to perform just as well as, or even better than, expert cytologists. Some notable methods for classifying cervix cancers include ResNet50, VGG16, MobileNetV2, and InceptionV3, based on deep convolutional neural networks (CNN). However, these methods are computationally expensive. We present CerviXpert, a multi-structural Convolutional Neural Network, to identify cervix cancer. We perform extensive experiments on a publicly available dataset, SiPaKMeD, to show the efficacy of our method. CerviXpert presents a promising solution for efficient cervical cancer screening and diagnosis by striking a balance between accuracy and practical feasibility.

In this study, combined DFT, SCAPS-1D, and wxAMPS frameworks are used to investigate the optimized designs of Cs2BiAgI6 double perovskite-based solar cells. The first-principle calculation is employed to investigate the structural stability, optical responses, and electronic contribution of the constituent elements in Cs2BiAgI6 absorber material, where SCAPS-1D and wxAMPS simulators are used to scrutinize different configurations of Cs2BiAgI6 solar cells. Here, PCBM, ZnO, TiO2, C60, IGZO, SnO2, WS2, and CeO2 are used as ETL, and Cu2O, CuSCN, CuSbS2, NiO, P3HT, PEDOT: PSS, Spiro-MeOTAD, CuI, CuO, V2O5, CBTS, CFTS are used as HTL, and Au is used as a back contact. About ninety-six combinations of Cs2BiAgI6-based solar cell structures are investigated, in which eight sets of solar cell structures are identified as the most efficient structures. Besides, holistic investigation on the effect of different factors such as the thickness of different layers, series and shunt resistances, temperature, capacitance, Mott-Schottky and generation-recombination rates, and J-V (current-voltage density) and QE (quantum efficiency) characteristics is performed. The results show CBTS as the best HTL for Cs2BiAgI6 with all eight ETLs used in this work, resulting in a power conversion efficiency (PCE) of 19.99%, 21.55%, 21.59%, 17.47%, 20.42%, 21.52%, 14.44%, 21.43% with PCBM, TiO2, ZnO, C60, IGZO, SnO2, CeO2, WS2, respectively. The proposed strategy may pave the way for further design optimization of lead-free double perovskite solar cells., Comment: 36 pages, 14 figures, 6 tables

This study has explored numerous physical properties of CaPd$_3$Ti$_4$O$_{1}$2 (CPTO) and CaPd$_3$V$_4$O$_{12}$ (CPVO) quadruple perovskites employing the density functional theory (DFT) method. The mechanical permanence of these two compounds was observed by the Born stability criteria as well. The band structure of CPTO reveals a 0.88 and 0.46 eV direct narrow band gap while using GGA-mBJ and GGA-PBE potentials, respectively, which is an indication of its fascinating semiconducting nature. The calculated partial density of states indicates the strong hybridization between Pd-4d and O-2p orbital electrons for CPTO, whereas Pd-4d and V-3d-O-2p for CPVO. The study of the chemical bonding nature and electronic charge distribution graph reveals the coexistence of covalent O-V/Pd bonds, ionic O-Ti/Ca bonds, as well as metallic Ti/V-Ti/V bonding for both compounds. The Fermi surface of CPVO ensures a kind of hole as well as electron faces simultaneously, indicating the multifarious band characteristic. The prediction of the static real dielectric function (optical property) of CPTO at zero energy implies its promising dielectric nature. The photoconductivity and absorption coefficient of CPBO display good qualitative compliance with the consequences of band structure computations. The calculated thermodynamic properties manifest the thermodynamical stability for CPBO, whereas phonon dispersions of CPVO exhibit stable phonon dispersion in contrast to slightly unstable phonon dispersion of CPTO. The predicted Debye temperature ($\theta_D$) has been utilized to correlate its topical features including thermoelectric behaviors. The studied thermoelectric transport properties of CPTO yielded the Seebeck coefficient (186 V/K), power factor (11.9 Wcm$^{-1}$K$^{-2}$), and figure of merit (ZT) value of about 0.8 at 800 K, indicating that this material could be a promising candidate for thermoelectric applications., Comment: 23 pages, 11 figures. arXiv admin note: substantial text overlap with arXiv:1904.05099

Early detection and modeling of a contagious epidemic can provide important guidance about quelling the contagion, controlling its spread, or the effective design of countermeasures. A topic of recent interest has been to design social network sensors, i.e., identifying a small set of people who can be monitored to provide insight into the emergence of an epidemic in a larger population. We formally pose the problem of designing social network sensors for flu epidemics and identify two different objectives that could be targeted in such sensor design problems. Using the graph theoretic notion of dominators we develop an efficient and effective heuristic for forecasting epidemics at lead time. Using six city-scale datasets generated by extensive microscopic epidemiological simulations involving millions of individuals, we illustrate the practical applicability of our methods and show significant benefits (up to twenty-two days more lead time) compared to other competitors. Most importantly, we demonstrate the use of surrogates or proxies for policy makers for designing social network sensors that require from nonintrusive knowledge of people to more information on the relationship among people. The results show that the more intrusive information we obtain, the longer lead time to predict the flu outbreak up to nine days., Comment: The conference version of the paper is submitted for publication

The Decision Support System (DSS) contains more than one antecedent and the degrees of strength of the antecedents need to be combined to determine the overall strength of the rule consequent. The membership values of the linguistic variables in Fuzzy have to be combined using an aggregation operator. But it is not feasible to predefine the form of aggregation operators in decision making. Instead, each rule should be found based on the feeling of the experts and on their actual decision pattern over the set of typical examples. Thus this work illustrates how the choice of aggregation operators is intended to mimic human decision making and can be selected and adjusted to fit empirical data, a series of test cases. Both parametrized and nonparametrized aggregation operators are adapted to fit empirical data. Moreover, they provided compensatory properties and, therefore, seemed to produce a better decision support system. To solve the problem, a threshold point from the output of the aggregation operators is chosen as the separation point between two classes. The best achieved accuracy is chosen as the appropriate aggregation operator. Thus a medical decision can be generated which is very close to a practitioner's guideline.

We report a helicity analysis of subthreshold rho production on 2H and 12C nuclei at low photoproduction energies and large |t|. The results are indicative of a large longitudinal rho polarization (l=1, m=0) and are consistent with a strong helicity-flip mechanism of rho production. The analysis is model-independent and supports the large mass shift reported from the earlier 3He experiments., Comment: 9 pages, 4 figures