Search

Interactive feedback, where feedback flows in both directions between teacher and student, is more effective than traditional one-way feedback. However, it is often too time-consuming for widespread use in educational practice. While Large Language Models (LLMs) have potential for automating feedback, they struggle with reasoning and interaction in an interactive setting. This paper introduces CAELF, a Contestable AI Empowered LLM Framework for automating interactive feedback. CAELF allows students to query, challenge, and clarify their feedback by integrating a multi-agent system with computational argumentation. Essays are first assessed by multiple Teaching-Assistant Agents (TA Agents), and then a Teacher Agent aggregates the evaluations through formal reasoning to generate feedback and grades. Students can further engage with the feedback to refine their understanding. A case study on 500 critical thinking essays with user studies demonstrates that CAELF significantly improves interactive feedback, enhancing the reasoning and interaction capabilities of LLMs. This approach offers a promising solution to overcoming the time and resource barriers that have limited the adoption of interactive feedback in educational settings.

This paper addresses the problem of end-to-end (E2E) design of learning and communication in a task-oriented semantic communication system. In particular, we consider a multi-device cooperative edge inference system over a wireless multiple-input multiple-output (MIMO) multiple access channel, where multiple devices transmit extracted features to a server to perform a classification task. We formulate the E2E design of feature encoding, MIMO precoding, and classification as a conditional mutual information maximization problem. However, it is notoriously difficult to design and train an E2E network that can be adaptive to both the task dataset and different channel realizations. Regarding network training, we propose a decoupled pretraining framework that separately trains the feature encoder and the MIMO precoder, with a maximum a posteriori (MAP) classifier employed at the server to generate the inference result. The feature encoder is pretrained exclusively using the task dataset, while the MIMO precoder is pretrained solely based on the channel and noise distributions. Nevertheless, we manage to align the pretraining objectives of each individual component with the E2E learning objective, so as to approach the performance bound of E2E learning. By leveraging the decoupled pretraining results for initialization, the E2E learning can be conducted with minimal training overhead. Regarding network architecture design, we develop two deep unfolded precoding networks that effectively incorporate the domain knowledge of the solution to the decoupled precoding problem. Simulation results on both the CIFAR-10 and ModelNet10 datasets verify that the proposed method achieves significantly higher classification accuracy compared to various baselines., Comment: major revision in IEEE JSAC

Embodied AI is transforming how AI systems interact with the physical world, yet existing datasets are inadequate for developing versatile, general-purpose agents. These limitations include a lack of standardized formats, insufficient data diversity, and inadequate data volume. To address these issues, we introduce ARIO (All Robots In One), a new data standard that enhances existing datasets by offering a unified data format, comprehensive sensory modalities, and a combination of real-world and simulated data. ARIO aims to improve the training of embodied AI agents, increasing their robustness and adaptability across various tasks and environments. Building upon the proposed new standard, we present a large-scale unified ARIO dataset, comprising approximately 3 million episodes collected from 258 series and 321,064 tasks. The ARIO standard and dataset represent a significant step towards bridging the gaps of existing data resources. By providing a cohesive framework for data collection and representation, ARIO paves the way for the development of more powerful and versatile embodied AI agents, capable of navigating and interacting with the physical world in increasingly complex and diverse ways. The project is available on https://imaei.github.io/project_pages/ario/, Comment: Project website: https://imaei.github.io/project_pages/ario/

We perform a comprehensive analysis of the minimal eclectic flavor group $Q_{8}\rtimes S_3$ which is isomorphic to $GL(2,3)$, and all its irreducible representations are induced from the irreducible representations of $Q_{8}$ and $S_{3}$. The consistency conditions between EFG and generalized CP (gCP) symmetry are revisited, and we find the gCP symmetry compatible with the minimal EFG $Q_{8}\rtimes S_3$. The most general forms of K\"ahler potential and superpotential based on $Q_{8}\rtimes S_3$ are discussed, and the corresponding fermion mass matrices are presented. A concrete lepton model invariant under $Q_{8}\rtimes S_3$ and gCP is constructed, in which the experimental data of all six lepton masses and six mixing parameters can be successfully described through seven real input parameters. The model predicts a vanishing effective mass $m_{\beta\beta}$ in neutrinoless double beta decay., Comment: 35 pages, 1 figures

Coherent elastic neutrino-nucleus scattering (CEvNS) provides a unique probe for neutrino properties Beyond the Standard Model (BSM) physics. REactor neutrino LIquid xenon Coherent Scattering experiment (RELICS), a proposed reactor neutrino program using liquid xenon time projection chamber (LXeTPC) technology, aims to investigate the CEvNS process of antineutrinos off xenon atomic nuclei. In this work, the design of the experiment is studied and optimized based on Monte Carlo (MC) simulations. To achieve a sufficiently low energy threshold for CEvNS detection, an ionization-only analysis channel is adopted for RELICS. A high emission rate of delayed electrons after a big ionization signal is the major background, leading to an analysis threshold of 120 photo-electrons in the CEvNS search. The second largest background, nuclear recoils induced by cosmic-ray neutrons, is suppressed via a passive water shield. The physics potential of RELICS is explored with a 32 kg-yr exposure at a baseline of 25 m from a reactor core with a 3 GW thermal power. In an energy range of 120 to 240 PE, we expect 4902.4 CEvNS and 1318.4 background events. The sensitivity of RELICS to the weak mixing angle is investigated at a low momentum transfer. Our study shows that RELICS can further improve the constraints on the non-standard neutrino interaction (NSI) compared to the current best results.

In task-oriented communications, most existing work designed the physical-layer communication modules and learning based codecs with distinct objectives: learning is targeted at accurate execution of specific tasks, while communication aims at optimizing conventional communication metrics, such as throughput maximization, delay minimization, or bit error rate minimization. The inconsistency between the design objectives may hinder the exploitation of the full benefits of task-oriented communications. In this paper, we consider a task-oriented multi-device edge inference system over a multiple-input multiple-output (MIMO) multiple-access channel, where the learning (i.e., feature encoding and classification) and communication (i.e., precoding) modules are designed with the same goal of inference accuracy maximization. Instead of end-to-end learning which involves both the task dataset and wireless channel during training, we advocate a separate design of learning and communication to achieve the consistent goal. Specifically, we leverage the maximal coding rate reduction (MCR2) objective as a surrogate to represent the inference accuracy, which allows us to explicitly formulate the precoding optimization problem. We cast valuable insights into this formulation and develop a block coordinate ascent (BCA) algorithm for efficient problem-solving. Moreover, the MCR2 objective serves the loss function for feature encoding and guides the classification design. Simulation results on the synthetic features explain the mechanism of MCR2 precoding at different SNRs. We also validate on the CIFAR-10 and ModelNet10 datasets that the proposed design achieves a better latency-accuracy tradeoff compared to various baselines. As such, our work paves the way for further exploration into the synergistic alignment of learning and communication objectives in task-oriented communication systems., Comment: under minor revision in IEEE Transactions on Wireless Communications

We have performed a systematical study of the eclectic flavor group $\Delta(27)\rtimes S_3$ which is the extension of the traditional flavor symmetry $\Delta(27)$ by the finite modular symmetry $S_3$. Consistency between $\Delta(27)$ and $S_3$ requires that the eight nontrivial singlet representations of $\Delta(27)$ should be arranged into four reducible doublets. The modular transformation matrices are determined for various $\Delta(27)$ multiplets, and the CP-like symmetry compatible with $\Delta(27)\rtimes S_3$ are discussed. We study the general form of the K\"ahler potential and superpotential invariant under $\Delta(27)\rtimes S_3$, and the corresponding fermion mass matrices are presented. We propose a bottom-up model for lepton masses and mixing based on $\Delta(27)\rtimes S_{3}$, a numerical analysis is performed and the experimental data can be accommodated., Comment: 36 pages, 2 figures

Cloud-native databases have become the de-facto choice for mission-critical applications on the cloud due to the need for high availability, resource elasticity, and cost efficiency. Meanwhile, driven by the increasing connectivity between data generation and analysis, users prefer a single database to efficiently process both OLTP and OLAP workloads, which enhances data freshness and reduces the complexity of data synchronization and the overall business cost. In this paper, we summarize five crucial design goals for a cloud-native HTAP database based on our experience and customers' feedback, i.e., transparency, competitive OLAP performance, minimal perturbation on OLTP workloads, high data freshness, and excellent resource elasticity. As our solution to realize these goals, we present PolarDB-IMCI, a cloud-native HTAP database system designed and deployed at Alibaba Cloud. Our evaluation results show that PolarDB-IMCI is able to handle HTAP efficiently on both experimental and production workloads; notably, it speeds up analytical queries up to $\times149$ on TPC-H (100 $GB$). PolarDB-IMCI introduces low visibility delay and little performance perturbation on OLTP workloads (< 5%), and resource elasticity can be achieved by scaling out in tens of seconds., Comment: 14 pages, 16 figures, to be published in ACM SIGMOD 2023

The K\"ahler potentials of modular symmetry models receive unsuppressed contributions which may be controlled by a flavor symmetry, where the combination of the two symmetry types is referred to as eclectic flavor symmetry. After briefly reviewing the consistency conditions of eclectic flavor symmetry models, including with generalised (g)CP, we perform a comprehensive bottom-up study of eclectic flavor symmetry models based on $\Omega(1)\cong \Delta(27)\rtimes T^\prime$, consisting of the flavor symmetry $\Delta(27)$ in a semi-direct product with the modular symmetry $T^\prime$. The modular transformations of different $\Delta(27)$ multiplets are given by solving the consistency condition. The eight nontrivial singlets of $\Delta(27)$ are related by $T'$ modular symmetry, and they have to be present or absent simultaneously in any $\Omega(1)$ model. The most general forms of the superpotential and K\"ahler potential invariant under $\Omega(1)$ are discussed, and the corresponding fermion mass matrices are presented. Based on the eclectic flavor group $\Omega(1)$, two concrete lepton models which can successfully describe the experimental data of lepton masses and mixing parameters are constructed. For the two models without gCP, all six mixing parameters vary in small regions. A nearly maximal atmospheric mixing angle $\theta_{23}$ and Dirac CP phase $\delta_{CP}$ are obtained in the first model. After considering the compatible gCP symmetry and the assumption of $\Re \tau=0$ in the first model, the $\mu-\tau$ reflection symmetry is preserved in the charged lepton diagonal basis. As a consequence, the atmospheric mixing angle and Dirac CP phase are predicted to be maximal, and two Majorana CP phases are predicted to be $\pi$., Comment: 47 pages, 2 figures

In next-generation wireless networks, reconfigurable intelligent surface (RIS)-assisted multiple-input multiple-output (MIMO) systems are foreseeable to support a large number of antennas at the transceiver as well as a large number of reflecting elements at the RIS. To fully unleash the potential of RIS, the phase shifts of RIS elements should be carefully designed, resulting in a high-dimensional non-convex optimization problem that is hard to solve. In this paper, we address this scalability issue by partitioning RIS into sub-surfaces, so as to optimize the phase shifts in sub-surface levels to reduce complexity. Specifically, each subsurface employs a linear phase variation structure to anomalously reflect the incident signal to a desired direction, and the sizes of sub-surfaces can be adaptively adjusted according to channel conditions. We formulate the achievable rate maximization problem by jointly optimizing the transmit covariance matrix and the RIS phase shifts. Under the RIS partitioning framework, the RIS phase shifts optimization reduces to the manipulation of the sub-surface sizes, the phase gradients of sub-surfaces, and the common phase shifts of sub-surfaces. Then, we characterize the asymptotic behavior of the system with an infinitely large number of transceiver antennas and RIS elements. The asymptotic analysis provides useful insights on the understanding of the fundamental performance-complexity tradeoff in RIS partitioning design. We show that in the asymptotic domain, the achievable rate maximization problem has a rather simple form. We develop an efficient algorithm to find an approximate optimal solution via a 1D grid search. By applying the asymptotic result to a finite-size system with necessary modifications, we show by numerical results that the proposed design achieves a favorable tradeoff between system performance and computational complexity., Comment: accepted by IEEE Transactions on Wireless Communications

Several problems in neuroimaging and beyond require inference on the parameters of multi-task sparse hierarchical regression models. Examples include M/EEG inverse problems, neural encoding models for task-based fMRI analyses, and climate science. In these domains, both the model parameters to be inferred and the measurement noise may exhibit a complex spatio-temporal structure. Existing work either neglects the temporal structure or leads to computationally demanding inference schemes. Overcoming these limitations, we devise a novel flexible hierarchical Bayesian framework within which the spatio-temporal dynamics of model parameters and noise are modeled to have Kronecker product covariance structure. Inference in our framework is based on majorization-minimization optimization and has guaranteed convergence properties. Our highly efficient algorithms exploit the intrinsic Riemannian geometry of temporal autocovariance matrices. For stationary dynamics described by Toeplitz matrices, the theory of circulant embeddings is employed. We prove convex bounding properties and derive update rules of the resulting algorithms. On both synthetic and real neural data from M/EEG, we demonstrate that our methods lead to improved performance., Comment: Accepted to the 35th Conference on Neural Information Processing Systems (NeurIPS 2021)

We propose to construct the finite modular groups from the quotient of two principal congruence subgroups as $\Gamma(N')/\Gamma(N")$, and the modular group $SL(2,\mathbb{Z})$ is extended to a principal congruence subgroup $\Gamma(N')$. The original modular invariant theory is reproduced when $N'=1$. We perform a comprehensive study of $\Gamma'_6$ modular symmetry corresponding to $N'=1$ and $N"=6$, five types of models for lepton masses and mixing with $\Gamma'_6$ modular symmetry are discussed and some example models are studied numerically. The case of $N'=2$ and $N"=6$ is considered, the finite modular group is $\Gamma(2)/\Gamma(6)\cong T'$, and a benchmark model is constructed., Comment: 39 pages, 4 figures

In reconfigurable intelligent surfaces (RISs) aided communications, the existing passive beamforming (PB) design involves polynomial complexity in the number of reflecting elements, and thus is difficult to implement due to a massive number of reflecting elements. To overcome this difficulty, we propose a reflection-angle-based cascaded channel model by adopting the generalized Snell's law, in which the dimension of the variable space involved in optimization is significantly reduced, resulting in a simplified hierarchical passive beamforming (HPB) design. We develop an efficient two-stage HPB algorithm, which exploits the angular domain property of the channel, to maximize the achievable rate of the target user. Simulation results demonstrate the appealing performance and low complexity of the proposed HPB design.

Electromagnetic source imaging (ESI) requires solving a highly ill-posed inverse problem. To seek a unique solution, traditional ESI methods impose various forms of priors that may not accurately reflect the actual source properties, which may hinder their broad applications. To overcome this limitation, in this paper a novel data-synthesized spatio-temporally convolutional encoder-decoder network method termed DST-CedNet is proposed for ESI. DST-CedNet recasts ESI as a machine learning problem, where discriminative learning and latent-space representations are integrated in a convolutional encoder-decoder network (CedNet) to learn a robust mapping from the measured electroencephalography/magnetoencephalography (E/MEG) signals to the brain activity. In particular, by incorporating prior knowledge regarding dynamical brain activities, a novel data synthesis strategy is devised to generate large-scale samples for effectively training CedNet. This stands in contrast to traditional ESI methods where the prior information is often enforced via constraints primarily aimed for mathematical convenience. Extensive numerical experiments as well as analysis of a real MEG and Epilepsy EEG dataset demonstrate that DST-CedNet outperforms several state-of-the-art ESI methods in robustly estimating source signals under a variety of source configurations., Comment: 15 pages, 14 figures, and journal

Annotating histopathological images is a time-consuming andlabor-intensive process, which requires broad-certificated pathologistscarefully examining large-scale whole-slide images from cells to tissues.Recent frontiers of transfer learning techniques have been widely investi-gated for image understanding tasks with limited annotations. However,when applied for the analytics of histology images, few of them can effec-tively avoid the performance degradation caused by the domain discrep-ancy between the source training dataset and the target dataset, suchas different tissues, staining appearances, and imaging devices. To thisend, we present a novel method for the unsupervised domain adaptationin histopathological image analysis, based on a backbone for embeddinginput images into a feature space, and a graph neural layer for propa-gating the supervision signals of images with labels. The graph model isset up by connecting every image with its close neighbors in the embed-ded feature space. Then graph neural network is employed to synthesizenew feature representation from every image. During the training stage,target samples with confident inferences are dynamically allocated withpseudo labels. The cross-entropy loss function is used to constrain thepredictions of source samples with manually marked labels and targetsamples with pseudo labels. Furthermore, the maximum mean diversityis adopted to facilitate the extraction of domain-invariant feature repre-sentations, and contrastive learning is exploited to enhance the categorydiscrimination of learned features. In experiments of the unsupervised do-main adaptation for histopathological image classification, our methodachieves state-of-the-art performance on four public datasets

Elastography ultrasound (EUS) provides additional bio-mechanical in-formation about lesion for B-mode ultrasound (BUS) in the diagnosis of breast cancers. However, joint utilization of both BUS and EUS is not popular due to the lack of EUS devices in rural hospitals, which arouses a novel modality im-balance problem in computer-aided diagnosis (CAD) for breast cancers. Current transfer learning (TL) pay little attention to this special issue of clinical modality imbalance, that is, the source domain (EUS modality) has fewer labeled samples than those in the target domain (BUS modality). Moreover, these TL methods cannot fully use the label information to explore the intrinsic relation between two modalities and then guide the promoted knowledge transfer. To this end, we propose a novel doubly supervised TL network (DDSTN) that integrates the Learning Using Privileged Information (LUPI) paradigm and the Maximum Mean Discrepancy (MMD) criterion into a unified deep TL framework. The proposed algorithm can not only make full use of the shared labels to effectively guide knowledge transfer by LUPI paradigm, but also perform additional super-vised transfer between unpaired data. We further introduce the MMD criterion to enhance the knowledge transfer. The experimental results on the breast ultra-sound dataset indicate that the proposed DDSTN outperforms all the compared state-of-the-art algorithms for the BUS-based CAD., Comment: Accepted by MICCAI 2020

The performance of a device-to-device (D2D) underlay communication system is limited by the co-channel interference between cellular users (CUs) and D2D devices. To address this challenge, an intelligent reflecting surface (IRS) aided D2D underlay system is studied in this paper. A two-timescale optimization scheme is proposed to reduce the required channel training and feedback overhead, where transmit beamforming at the base station (BS) and power control at the D2D transmitter are adapted to instantaneous effective channel state information (CSI); and the IRS phase shifts are adapted to slow-varying channel mean. Based on the two-timescale optimization scheme, we aim to maximize the D2D ergodic rate subject to a given outage probability constrained signal-to-interference-plus-noise ratio (SINR) target for the CU. The two-timescale problem is decoupled into two sub-problems, and the two sub-problems are solved iteratively with closed-form expressions. Numerical results verify that the two-timescale based optimization performs better than several baselines, and also demonstrate a favorable trade-off between system performance and CSI overhead.

We consider for the first time level 7 modular invariant flavour models where the lepton mixing originates from the breaking of modular symmetry and couplings responsible for lepton masses are modular forms. The latter are decomposed into irreducible multiplets of the finite modular group $\Gamma_7$, which is isomorphic to $PSL(2,Z_{7})$, the projective special linear group of two dimensional matrices over the finite Galois field of seven elements, containing 168 elements, sometimes written as $PSL_2(7)$ or $\Sigma(168)$. At weight 2, there are 26 linearly independent modular forms, organised into a triplet, a septet and two octets of $\Gamma_7$. A full list of modular forms up to weight 8 are provided. Assuming the absence of flavons, the simplest modular-invariant models based on $\Gamma_7$ are constructed, in which neutrinos gain masses via either the Weinberg operator or the type-I seesaw mechanism, and their predictions compared to experiment., Comment: 45 pages, 3 figures