Search

Enhancing light emission from perovskite nanocrystal (NC) films is essential in light-emitting devices, as their conventional stacks often restrict the escape of emitted light. This work addresses this challenge by employing a TiO$_2$ grating to enhance light extraction and shape the emission of CsPbBr$_3$ nanocrystal films. Angle-resolved photoluminescence (PL) demonstrated a tenfold increase in emission intensity by coupling the Bloch resonances of the grating with the spontaneous emission of the perovskite NCs. Fluorescence lifetime imaging microscopy (FLIM) provided micrometer-resolution mapping of both PL intensity and lifetime across a large area, revealing a decrease in PL lifetime from 8.2 ns for NC films on glass to 6.1 ns on the TiO$_2$ grating. Back focal plane (BFP) spectroscopy confirmed how the Bloch resonances transformed the unpolarized, spatially incoherent emission of NCs into polarized and directed light. These findings provide further insights into the interactions between dielectric nanostructures and perovskite NC films, offering possible pathways for designing better performing perovskite optoelectronic devices.

Domain specific question answering is an evolving field that requires specialized solutions to address unique challenges. In this paper, we show that a hybrid approach combining a fine-tuned dense retriever with keyword based sparse search methods significantly enhances performance. Our system leverages a linear combination of relevance signals, including cosine similarity from dense retrieval, BM25 scores, and URL host matching, each with tunable boost parameters. Experimental results indicate that this hybrid method outperforms our single-retriever system, achieving improved accuracy while maintaining robust contextual grounding. These findings suggest that integrating multiple retrieval methodologies with weighted scoring effectively addresses the complexities of domain specific question answering in enterprise settings.

Let $L$ be the Dunkl Laplacian on the Euclidean space $\mathbb R^N$ associated with a normalized root $R$ and a multiplicity function $k(\nu)\ge 0, \nu\in R$. In this paper, we first prove that the Besov and Triebel-Lizorkin spaces associated with the Dunkl Laplacian $L$ are identical to the Besov and Triebel-Lizorkin spaces defined in the space of homogeneous type $(\mathbb R^N, \|\cdot\|, dw)$, where $dw({\rm x})=\prod_{\nu\in R}\langle \nu,{\rm x}\rangle^{k(\nu)}d{\rm x}$. Next, consider the Dunkl transform denoted by $\mathcal{F}$. We introduce the multiplier operator $T_m$, defined as $T_mf = \mathcal{F}^{-1}(m\mathcal{F}f)$, where $m$ is a bounded function defined on $\mathbb{R}^N$. Our second aim is to prove multiplier theorems, including the H\"ormander multiplier theorem, for $T_m$ on the Besov and Tribel-Lizorkin spaces in the space of homogeneous type $(\mathbb R^N, \|\cdot\|, dw)$. Importantly, our findings present novel results, even in the specific case of the Hardy spaces., Comment: 57 pages. See also at: https://www.researchgate.net/publication/371952480_Harmonic_analysis_in_Dunkl_settings

Traditional deep learning methods in medical imaging often focus solely on segmentation or classification, limiting their ability to leverage shared information. Multi-task learning (MTL) addresses this by combining both tasks through shared representations but often struggles to balance local spatial features for segmentation and global semantic features for classification, leading to suboptimal performance. In this paper, we propose a simple yet effective UNet-based MTL model, where features extracted by the encoder are used to predict classification labels, while the decoder produces the segmentation mask. The model introduces an advanced encoder incorporating a novel ResFormer block that integrates local context from convolutional feature extraction with long-range dependencies modeled by the Transformer. This design captures broader contextual relationships and fine-grained details, improving classification and segmentation accuracy. To enhance classification performance, multi-scale features from different encoder levels are combined to leverage the hierarchical representation of the input image. For segmentation, the features passed to the decoder via skip connections are refined using a novel dilated feature enhancement (DFE) module, which captures information at different scales through three parallel convolution branches with varying dilation rates. This allows the decoder to detect lesions of varying sizes with greater accuracy. Experimental results across multiple medical datasets confirm the superior performance of our model in both segmentation and classification tasks, compared to state-of-the-art single-task and multi-task learning methods.

Let $\nu=(\nu_1,\ldots,\nu_n)\in (-1,\infty)^n$, $n\ge 1$, and let $\mathcal{L}_\nu$ be a self-adjoint extension of the differential operator \[ L_\nu := \sum_{i=1}^n \left[-\frac{\partial^2}{\partial x_i^2} + x_i^2 + \frac{1}{x_i^2}(\nu_i^2 - \frac{1}{4})\right] \] on $C_c^\infty(\mathbb{R}_+^n)$ as the natural domain. In this paper, we investigate the weighted estimates of singular integrals in the Laguerre setting including the maximal function, the Riesz transform and the square functions associated to the Laguerre operator $\mathcal L_\nu$. In the special case of the Riesz transform, the paper completes the description of the Riesz transform for the full range of $\nu\in (-1,\infty)^n$ which significant improves the result in [J. Funct. Anal. 244 (2007), 399--443] for $\nu_i\ge -1/2, \nu_i\notin (-1/2,1/2)$ for $i=1,\ldots, n$., Comment: 31 pages

Let $\nu\in [-1/2,\infty)^n$, $n\ge 1$, and let $\mathcal{L}_\nu$ be a self-adjoint extension of the differential operator \[ L_\nu := \sum_{i=1}^n \left[-\frac{\partial^2}{\partial x_i^2} + x_i^2 + \frac{1}{x_i^2}(\nu_i^2 - \frac{1}{4})\right] \] on $C_c^\infty(\mathbb{R}_+^n)$ as the natural domain. In this paper, we first prove that the Riesz transform associated with $\mathcal L_\nu$ is a Calder\'on-Zygmund operator, answering the open problem in [JFA, 244 (2007), 399-443]. In addition, we develop the theory of Hardy spaces and Campanato spaces associated with $\mathcal{L}_\nu$. As applications, we prove that the Riesz transform related to $\mathcal{L}_\nu$ is bounded on these Hardy spaces and Campanato spaces, completing the description of the boundedness of the Riesz transform in the Laguerre expansion setting., Comment: 48 pages

Consider the discrete Laplacian $\Delta_d$ defined on the set of integers $\mathbb Z$ by \[ \Delta_d f(n) = -f(n+1) + 2f(n) -f(n-1), \ \ \ \ n\in \mathbb Z, \] where $f$ is a function defined on $\mathbb Z$. In this paper, we define Hardy spaces, Besov spaces and Triebel--Lizorkin spaces associated with $\Delta_d$ and then show that these function spaces coincide with the classical function spaces defined on $\mathbb Z$. As applications, we prove the boundedness of the spectral multipliers and the Riesz transforms associated with $\Delta_d$ on these function spaces., Comment: 49 pages. Accepted to publish in Annali Scuola Normale Speriore - Classe de Scienze in 2023

Accurate identification of software vulnerabilities is crucial for system integrity. Vulnerability datasets, often derived from the National Vulnerability Database (NVD) or directly from GitHub, are essential for training machine learning models to detect these security flaws. However, these datasets frequently suffer from significant noise, typically 40% to 75%, due primarily to the automatic and indiscriminate labeling of all changes in vulnerability-fixing commits (VFCs) as vulnerability-related. This misclassification occurs because not all changes in a commit aimed at fixing vulnerabilities pertain to security threats; many are routine updates like bug fixes or test improvements. This paper introduces the first methodology that uses the Large Language Model (LLM) with a heuristic enhancement to automatically identify vulnerability-fixing changes from VFCs, achieving an F1-score of 0.82. VulSifter was applied to a large-scale study, where we conducted a crawl of 127,063 repositories on GitHub, resulting in the acquisition of 5,352,105 commits. VulSifter involves utilizing an LLM to comprehend code semantics and contextual information, while applying heuristics to filter out unrelated changes. We then developed CleanVul, a high-quality dataset comprising 11,632 functions using our LLM heuristic enhancement approach, demonstrating Correctness (90.6%) comparable to established datasets such as SVEN and PrimeVul. To evaluate the CleanVul dataset, we conducted experiments focusing on fine-tuning various LLMs on CleanVul and other high-quality datasets. Evaluation results reveal that LLMs fine-tuned on CleanVul not only exhibit enhanced accuracy but also superior generalization capabilities compared to those trained on uncleaned datasets. Specifically, models trained on CleanVul and tested on PrimeVul achieve accuracy higher than those trained and tested exclusively on PrimeVul.

It has recently been shown that Terahertz sensors can effectively detect the spin resonances of Dirac fermions in graphene. The associated photovoltaic measurement technique allows for the investigation of the intrinsic spin-orbit coupling in graphene as well as its topological properties from microwave to Terahertz frequencies. In this work, using graphene/MoS2-based Field-Effect Transistors, we observed a magnetic resonance photovoltage signal in the Gigahertz range that is independent of the gate bias. The dispersion of the associated spin-flip transitions remains intriguingly unaffected by the MoS2 layer. In parallel, the spin-related signal consistently appears as a drop in photovoltage, regardless of the signal's polarity or origin, whether it arises from plasma wave rectification or thermoelectric effects. This behavior is interpreted as a decrease in the system's spin polarization due to spin-dependent recombination or scattering of photocarriers. Understanding the various photovoltaic signals in highly sensitive Gigahertz/Terahertz sensors paves the way for exploring spin-dependent mechanisms in two-dimensional quantum materials, influenced by proximity effects such as spin-orbit coupling, topology, and magnetism., Comment: 23 pages, 6 figures in the main text and 6 in the supplementary materials

Developing an interpretable system for generating reports in chest X-ray (CXR) analysis is becoming increasingly crucial in Computer-aided Diagnosis (CAD) systems, enabling radiologists to comprehend the decisions made by these systems. Despite the growth of diverse datasets and methods focusing on report generation, there remains a notable gap in how closely these models' generated reports align with the interpretations of real radiologists. In this study, we tackle this challenge by initially introducing Fine-Grained CXR (FG-CXR) dataset, which provides fine-grained paired information between the captions generated by radiologists and the corresponding gaze attention heatmaps for each anatomy. Unlike existing datasets that include a raw sequence of gaze alongside a report, with significant misalignment between gaze location and report content, our FG-CXR dataset offers a more grained alignment between gaze attention and diagnosis transcript. Furthermore, our analysis reveals that simply applying black-box image captioning methods to generate reports cannot adequately explain which information in CXR is utilized and how long needs to attend to accurately generate reports. Consequently, we propose a novel explainable radiologist's attention generator network (Gen-XAI) that mimics the diagnosis process of radiologists, explicitly constraining its output to closely align with both radiologist's gaze attention and transcript. Finally, we perform extensive experiments to illustrate the effectiveness of our method. Our datasets and checkpoint is available at https://github.com/UARK-AICV/FG-CXR., Comment: ACCV 2024

There has been a growing excitement that implicit graph generative models could be used to design or discover new molecules for medicine or material design. Because these molecules have not been discovered, they naturally lie in unexplored or scarcely supported regions of the distribution of known molecules. However, prior evaluation methods for implicit graph generative models have focused on validating statistics computed from the thick support (e.g., mean and variance of a graph property). Therefore, there is a mismatch between the goal of generating novel graphs and the evaluation methods. To address this evaluation gap, we design a novel evaluation method called Vertical Validation (VV) that systematically creates thin support regions during the train-test splitting procedure and then reweights generated samples so that they can be compared to the held-out test data. This procedure can be seen as a generalization of the standard train-test procedure except that the splits are dependent on sample features. We demonstrate that our method can be used to perform model selection if performance on thin support regions is the desired goal. As a side benefit, we also show that our approach can better detect overfitting as exemplified by memorization., Comment: Accepted to UAI 2024

We study the maximum set coverage problem in the massively parallel model. In this setting, $m$ sets that are subsets of a universe of $n$ elements are distributed among $m$ machines. In each round, these machines can communicate with each other, subject to the memory constraint that no machine may use more than $\tilde{O}(n)$ memory. The objective is to find the $k$ sets whose coverage is maximized. We consider the regime where $k = \Omega(m)$, $m = O(n)$, and each machine has $\tilde{O}(n)$ memory. Maximum coverage is a special case of the submodular maximization problem subject to a cardinality constraint. This problem can be approximated to within a $1-1/e$ factor using the greedy algorithm, but this approach is not directly applicable to parallel and distributed models. When $k = \Omega(m)$, to obtain a $1-1/e-\epsilon$ approximation, previous work either requires $\tilde{O}(mn)$ memory per machine which is not interesting compared to the trivial algorithm that sends the entire input to a single machine, or requires $2^{O(1/\epsilon)} n$ memory per machine which is prohibitively expensive even for a moderately small value $\epsilon$. Our result is a randomized $(1-1/e-\epsilon)$-approximation algorithm that uses $O(1/\epsilon^3 \cdot \log m \cdot (\log (1/\epsilon) + \log m))$ rounds. Our algorithm involves solving a slightly transformed linear program of the maximum coverage problem using the multiplicative weights update method, classic techniques in parallel computing such as parallel prefix, and various combinatorial arguments.

Background: Beta-delayed proton emission from neutron halo nuclei $^{11}$Be represents a rare decay process. The existence of the narrow resonance near the proton-emission threshold in $^{11}$B explains its unexpectedly high probability. However, the accurate value of the branching ratio remains challenging to determine. Purpose: We aim to provide a microscopic potential model to determine the branching ratio for beta-delayed proton emission from $^{11}$Be. We focus on quantifying the influence of the narrow resonance near the proton emission threshold on the result of the branching ratio. Method: We employ the Skyrme Hartree-Fock calculation within the potential model to obtain the branching ratio. We derive the single-particle potentials for the halo neutron and the emitting proton from the Skyrme Hartree-Fock calculation with minimal adjustment. As the resonance position is tightly linked to the potential depth, we can demonstrate quantitatively how variations in its location impact the outcome. Result: Slight variations in the resonance position significantly impact the branching ratio, with the upper limit reaching the order of $10^{-5}$. Conclusion: Experimental determination of the resonance energy, particularly whether it lies below $200$ keV, is crucial for determining the value of the branching ratio., Comment: 6 pages, 3 figures, submitted to Physical Review C

In this paper, we derive first and second-order optimality conditions of KKT type for locally optimal solutions to a class of multiobjective optimal control problems with endpoint constraint and mixed pointwise constraints. We give some sufficient conditions for normality of multipliers. Namely, we show that if the linearized system is controllable or some constraint qualifications are satisfied, then the multiplier corresponding to the objective function is different from zero. To demonstrate the practical relevance of our theoretical results, we apply these conditions to a multiobjective optimal control problem for sustainable energy management in smart grids, providing insights into the trade-offs between cost, renewable energy utilization, environmental impact, and grid stability., Comment: 27 pages

While small language models (SLMs) show promises for mobile deployment, their real-world performance and applications on smartphones remains underexplored. We present SlimLM, a series of SLMs optimized for document assistance tasks on mobile devices. Through extensive experiments on a Samsung Galaxy S24, we identify the optimal trade-offs between model size (ranging from 125M to 7B parameters), context length, and inference time for efficient on-device processing. SlimLM is pre-trained on SlimPajama-627B and fine-tuned on DocAssist, our constructed dataset for summarization, question answering and suggestion tasks. Our smallest model demonstrates efficient performance on S24, while larger variants offer enhanced capabilities within mobile constraints. We evaluate SlimLM against existing SLMs, showing comparable or superior performance and offering a benchmark for future research in on-device language models. We also provide an Android application, offering practical insights into SLM deployment. Our findings provide valuable insights and illuminate the capabilities of running advanced language models on high-end smartphones, potentially reducing server costs and enhancing privacy through on-device processing.

Training a high-quality Federated Learning (FL) model at the network edge is challenged by limited transmission resources. Although various device scheduling strategies have been proposed, it remains unclear how scheduling decisions affect the FL model performance under temporal constraints. This is pronounced when the wireless medium is shared to enable the participation of heterogeneous Internet of Things (IoT) devices with distinct communication modes: (1) a scheduling (pull) scheme, that selects devices with valuable updates, and (2) random access (push), in which interested devices transmit model parameters. The motivation for pushing data is the improved representation of own data distribution within the trained FL model and thereby better generalization. The scheduling strategy affects the transmission opportunities for push-based communication during the access phase, extending the number of communication rounds required for model convergence. This work investigates the interplay of push-pull interactions in a time-constrained FL setting, where the communication opportunities are finite, with a utility-based analytical model. Using real-world datasets, we provide a performance tradeoff analysis that validates the significance of strategic device scheduling under push-pull wireless access for several practical settings. The simulation results elucidate the impact of the device sampling strategy on learning efficiency under timing constraints.

A class of time-optimal control problems governed by semilinear parabolic equations with mixed pointwise constraints and final point constraints is considered. By introducing the so-called locally optimal solution to time-optimal control problems, we establish first and second-order necessary optimality conditions of KKT-type and second-order sufficient conditions for locally optimal solutions to the problem., Comment: 38 pages

Let $L$ be a closed, densely defined operator on $L^2(\mathbb{R}^n)$ satisfying suitable $L^p-L^q$ off-diagonal estimates of order $\kappa > 0$. This paper aims to investigate the two-weight estimate and the Bloom weighted estimate for the fractional operator $L^{-\alpha/\kappa}$ with $0 < \alpha < n$ through the method of sparse domination. Our assumptions on the operators are minimal, and our result applies to a wide range of differential operators. As a byproduct, we also establish a new sparse domination criterion for a general class of fractional operators, including the classical fractional integral.

By leveraging the representation power of deep neural networks, neural upper confidence bound (UCB) algorithms have shown success in contextual bandits. To further balance the exploration and exploitation, we propose Neural-$\sigma^2$-LinearUCB, a variance-aware algorithm that utilizes $\sigma^2_t$, i.e., an upper bound of the reward noise variance at round $t$, to enhance the uncertainty quantification quality of the UCB, resulting in a regret performance improvement. We provide an oracle version for our algorithm characterized by an oracle variance upper bound $\sigma^2_t$ and a practical version with a novel estimation for this variance bound. Theoretically, we provide rigorous regret analysis for both versions and prove that our oracle algorithm achieves a better regret guarantee than other neural-UCB algorithms in the neural contextual bandits setting. Empirically, our practical method enjoys a similar computational efficiency, while outperforming state-of-the-art techniques by having a better calibration and lower regret across multiple standard settings, including on the synthetic, UCI, MNIST, and CIFAR-10 datasets.

Technical debt (TD) is a term used to describe the additional work and costs that emerge when developers have opted for a quick and easy solution to a problem, rather than a more effective and well-designed, but time-consuming approach. Self-Admitted Technical Debts (SATDs) are a specific type of technical debts that developers intentionally document and acknowledge, typically via textual comments. While these self-admitted comments are a useful tool for identifying technical debts, most of the existing approaches focus on capturing crucial tokens associated with various categories of TD, neglecting the rich information embedded within the source code itself. Recent research has focused on detecting SATDs by analyzing comments embedded in source code, and there has been little work dealing with technical debts contained in the source code. To fill such a gap, in this study, through the analysis of comments and their associated source code from 974 Java projects hosted in the Stack corpus, we curated the first ever dataset of TD identified by code comments, coupled with its associated source code. Through an empirical evaluation, we found out that the comments of the resulting dataset help enhance the prediction performance of state-of-the-art SATD detection models. More importantly, including the classified source code significantly improves the accuracy in predicting various types of technical debt. In this respect, our work is two-fold: (i) We believe that our dataset will catalyze future work in the domain, inspiring various research issues related to the recognition of technical debt; (ii) The proposed classifiers may serve as baselines for other studies on the detection of TD by means of the curated dataset., Comment: The paper has been submitted to the Transactions on Software Engineering, and is now under review

The quantum mechanics of three interacting particles gives rise to interesting universal phenomena, such as the staircase of Efimov trimers predicted in the context of nuclear physics and observed in ultracold gases. Here, we observe a novel type of halo trimer using radiofrequency spectroscopy in an ultracold mixture of $^{23}$Na and $^{40}$K atoms. The trimers consist of two light bosons and one heavy fermion, and have the structure of a Feshbach dimer weakly bound to one additional boson. We find that the trimer peak closely follows the dimer resonance over the entire range of explored interaction strengths across an order of magnitude variation of the dimer energy, as reproduced by our theoretical analysis. The presence of this halo trimer is of direct relevance for many-body physics in ultracold mixtures and the association of ultracold molecules., Comment: 9+3 pages, 7+1 figures

Warning: this paper contains content that may be offensive or upsetting Hate speech moderation on global platforms poses unique challenges due to the multimodal and multilingual nature of content, along with the varying cultural perceptions. How well do current vision-language models (VLMs) navigate these nuances? To investigate this, we create the first multimodal and multilingual parallel hate speech dataset, annotated by a multicultural set of annotators, called Multi3Hate. It contains 300 parallel meme samples across 5 languages: English, German, Spanish, Hindi, and Mandarin. We demonstrate that cultural background significantly affects multimodal hate speech annotation in our dataset. The average pairwise agreement among countries is just 74%, significantly lower than that of randomly selected annotator groups. Our qualitative analysis indicates that the lowest pairwise label agreement-only 67% between the USA and India-can be attributed to cultural factors. We then conduct experiments with 5 large VLMs in a zero-shot setting, finding that these models align more closely with annotations from the US than with those from other cultures, even when the memes and prompts are presented in the dominant language of the other culture. Code and dataset are available at https://github.com/MinhDucBui/Multi3Hate., Comment: Preprint

Existing LLM agent systems typically select actions from a fixed and predefined set at every step. While this approach is effective in closed, narrowly-scoped environments, we argue that it presents two major challenges when deploying LLM agents in real-world scenarios: (1) selecting from a fixed set of actions significantly restricts the planning and acting capabilities of LLM agents, and (2) this approach requires substantial human effort to enumerate and implement all possible actions, which becomes impractical in complex environments with a vast number of potential actions. In this work, we propose an LLM agent framework that enables the dynamic creation and composition of actions in an online manner. In this framework, the agent interacts with the environment by generating and executing programs written in a general-purpose programming language at each step. Furthermore, generated actions are accumulated over time for future reuse. Our extensive experiments on the GAIA benchmark demonstrate that this framework offers significantly greater flexibility and outperforms previous methods. Notably, it allows an LLM agent to recover in scenarios where no relevant action exists in the predefined set or when existing actions fail due to unforeseen edge cases. At the time of writing, we hold the top position on the GAIA public leaderboard. Our code can be found in \href{https://github.com/adobe-research/dynasaur}{https://github.com/adobe-research/dynasaur}., Comment: 15 pages, 8 figures

Predicting program behavior and reasoning about code execution remain significant challenges in software engineering, particularly for large language models (LLMs) designed for code analysis. While these models excel at understanding static syntax, they often struggle with dynamic reasoning tasks. We introduce Visual Coder, a simple yet effective approach that enhances code reasoning by integrating multimodal Chain-of-Thought (CoT) reasoning with a visual Control Flow Graph (CFG). By aligning code snippets with their corresponding CFGs, Visual Coder provides deeper insights into execution flow, enabling more accurate predictions of code behavior. Our experiments demonstrate that augmenting LLMs with visual CFGs significantly outperforms text-based CFG descriptions in code reasoning tasks. We address challenges in multimodal CoT integration through a reference mechanism, ensuring consistency between code and its execution path, thereby improving performance in program behavior prediction, error detection, and output generation.

Accurately quantifying uncertainty in large language models (LLMs) is crucial for their reliable deployment, especially in high-stakes applications. Current state-of-the-art methods for measuring semantic uncertainty in LLMs rely on strict bidirectional entailment criteria between multiple generated responses and also depend on sequence likelihoods. While effective, these approaches often overestimate uncertainty due to their sensitivity to minor wording differences, additional correct information, and non-important words in the sequence. We propose a novel approach that leverages semantic embeddings to achieve smoother and more robust estimation of semantic uncertainty in LLMs. By capturing semantic similarities without depending on sequence likelihoods, our method inherently reduces any biases introduced by irrelevant words in the answers. Furthermore, we introduce an amortised version of our approach by explicitly modelling semantics as latent variables in a joint probabilistic model. This allows for uncertainty estimation in the embedding space with a single forward pass, significantly reducing computational overhead compared to existing multi-pass methods. Experiments across multiple question-answering datasets and frontier LLMs demonstrate that our embedding-based methods provide more accurate and nuanced uncertainty quantification than traditional approaches.

We present a multi-probe transport analysis that effectively separates bulk and edge currents in large Hall bar devices with standard geometries. Applied to transport measurements on all possible four-probe configurations of six-probe Hall bar devices made of inverted three-layer InAs/GaInSb quantum wells (QWs), our analysis not only reveals the presence of dissipative edge currents in the topological gap, but also allows the temperature dependence of bulk and edge conductivity to be evaluated separately. The temperature dependence of the edge conductivity for Hall bar channels from 10 $\mu$m to 70~$\mu$m in the range of 1.5 K to 45 K is consistent with the theoretical expectation of weakly interacting helical edge electrons with backscattering due to localized magnetic moments of charge impurities. We argue that these charge impurities are naturally associated with intrinsic Ga-antisite defects, which act as double acceptors in InAs/Ga(In)Sb-based QWs.

Non-Gaussian likelihoods are essential for modelling complex real-world observations but pose significant computational challenges in learning and inference. Even with Gaussian priors, non-Gaussian likelihoods often lead to analytically intractable posteriors, necessitating approximation methods. To this end, we propose efficient schemes to approximate the effects of non-Gaussian likelihoods by Gaussian densities based on variational inference and moment matching in transformed bases. These enable efficient inference strategies originally designed for models with a Gaussian likelihood to be deployed. Our empirical results demonstrate that the proposed matching strategies attain good approximation quality for binary and multiclass classification in large-scale point-estimate and distributional inferential settings. In challenging streaming problems, the proposed methods outperform all existing likelihood approximations and approximate inference methods in the exact models. As a by-product, we show that the proposed approximate log-likelihoods are a superior alternative to least-squares on raw labels for neural network classification.

Thermoelectricity in telluride materials is often improved by replacing telluride with selenium in its crystal. Most work, however, focuses on bulk crystal and leaves the 2D thin films intact. In this paper, we optimize the fabrication of selenium-doped bismuth telluride (Bi$_2$Te$_{3-\rm{x}}$Se$_{\rm{x}}$) thin films using a 3-source thermal co-evaporation. Thermoelectric properties, including the Seebeck coefficient and electrical resistivity, are systematically characterized to evaluate the material's performance for thermoelectric applications near room temperature. The thin films were deposited under carefully controlled conditions, with the evaporation rates of bismuth, tellurium, and selenium precisely monitored to achieve the desired stoichiometry and crystalline phase. Finally, thermoelectricity in Bi$_2$Te$_{3-\rm{x}}$Se$_{\rm{x}}$ at the ultra-thin regime is investigated. We consistently obtain films with thickness near 30 nm with a Seebeck coefficient of 400 $\mu$V/K and a power factor of 1 mW/mK$^2$.

Autoencoder permits the end-to-end optimization and design of wireless communication systems to be more beneficial than traditional signal processing. However, this emerging learning-based framework has weaknesses, especially sensitivity to physical attacks. This paper explores adversarial attacks against a double reconfigurable intelligent surface (RIS)-assisted multiple-input and multiple-output (MIMO)-based autoencoder, where an adversary employs encoded and decoded datasets to create adversarial perturbation and fool the system. Because of the complex and dynamic data structures, adversarial attacks are not unique, each having its own benefits. We, therefore, propose three algorithms generating adversarial examples and perturbations to attack the RIS-MIMO-based autoencoder, exploiting the gradient descent and allowing for flexibility via varying the input dimensions. Numerical results show that the proposed adversarial attack-based algorithm significantly degrades the system performance regarding the symbol error rate compared to the jamming attacks., Comment: 5 pages, 2 figures. Accepted by WCL

We study some properties of the growth rate of $\mathcal{L}(\mathcal{A},\mathcal{F})$, that is, the language of words over the alphabet $\mathcal{A}$ avoiding the set of forbidden factors $\mathcal{F}$. We first provide a sufficient condition on $\mathcal{F}$ and $\mathcal{A}$ for the growth of $\mathcal{L}(\mathcal{A},\mathcal{F})$ to be boundedly supermultiplicative. That is, there exist constants $C>0$ and $\alpha\ge0$, such that for all $n$, the number of words of length $n$ in $\mathcal{L}(\mathcal{A},\mathcal{F})$ is between $\alpha^n$ and $C\alpha^n$. In some settings, our condition provides a way to compute $C$, which implies that $\alpha$, the growth rate of the language, is also computable whenever our condition holds. We also apply our technique to the specific setting of power-free words where the argument can be slightly refined to provide better bounds. Finally, we apply a similar idea to $\mathcal{F}$-free circular words and in particular we make progress toward a conjecture of Shur about the number of square-free circular words.

Efficient long-context language modeling remains a significant challenge in Natural Language Processing (NLP). While Transformers dominate language tasks, they struggle with long sequences due to quadratic computational complexity in training and linearly scaling memory costs during inference. Recent State Space Models (SSMs) such as Mamba offer alternatives with constant memory usage, but they underperform in tasks requiring extensive in-context retrieval. We introduce Taipan, a novel hybrid architecture that combines Mamba-2 with Selective Attention Layers (SALs). These SALs identify tokens requiring long-range interactions, remove less important features, and then augment their representations using the attention module. This approach balances Mamba's efficiency with Transformer-like performance in memory-intensive tasks. By constraining the attention budget, Taipan extends accurate predictions to context lengths of up to 1 million tokens while preserving computational efficiency. Our experiments demonstrate Taipan's superior performance across various scales and tasks, offering a promising solution for efficient long-context language modeling.

Generalizing some popular sequences like Catalan's number, Schr\"oder's number, etc, we consider the sequence $s_n$ with $s_0=1$ and for $n\ge 1$, \begin{multline*} s_n=\sum_{x_1+\dots+x_{\ell_1}=n-1} \kappa_1 s_{x_1}\dots s_{x_{\ell_1}} + \dots +\sum_{x_1+\dots+x_{\ell_{t'}}=n-1} \kappa_{t'} s_{x_1}\dots s_{x_{\ell_{t'}}}+\\ \max_{x_1+\dots+x_{\ell_{t'+1}}=n-1} \kappa_{t'+1} s_{x_1}\dots s_{x_{\ell_{t'+1}}} + \dots + \max_{x_1+\dots+x_{\ell_t}=n-1} \kappa_t s_{x_1}\dots s_{x_{\ell_t}}, \end{multline*} where $x_i$ are nonnegative integers, $\ell_1,\dots,\ell_t$ are positive integers, and $\kappa_1,\dots,\kappa_t$ are positive reals. We show that it is possible to compute the growth rate $\lambda$ of $s_n$ to any precision. In particular, for every $n\ge 2$, \[ \sqrt[n]{\frac{\kappa^*}{\mathcal L(n-1) s_1} s_n} \le \lambda \le \sqrt[n]{3^{18\log 3 + 2\log\frac{s_1\mathcal L^2}{\kappa^*}} n^{3\log n + 12\log 3 + \log\frac{s_1\mathcal L^2}{\kappa^*}} s_n}, \]where $\mathcal L=\max_i \ell_i$ and $\kappa^*=\kappa_i$ for some $i$ with $\ell_i\ge 2$, and the logarithm has the base $\frac{\mathcal L+1}{\mathcal L}$. The constants in the inequalities are not very well optimized and serve mostly as a proof of concept with the ratio of the upper bound and the lower bound converging to $1$ as $n$ goes to infinity., Comment: 8 pages; comments are welcome

Entanglement breaking (EB) channels, as completely positive and trace-preserving linear operators, sever the entanglement between the input system and other systems. In the realm of infinite-dimensional systems, a related concept known as strongly entanglement breaking (SEB) channels emerges. This paper delves into characterizations of SEB channels, delineating necessary and sufficient conditions for a channel to be classified as SEB, especially with respect to the commutativity of its range. Moreover, we demonstrate that every closed self-adjoint subspace of trace-zero operators, with the trace norm, is the null space of a SEB channel., Comment: 12 pages

Diffusion models excel at generating visually striking content from text but can inadvertently produce undesirable or harmful content when trained on unfiltered internet data. A practical solution is to selectively removing target concepts from the model, but this may impact the remaining concepts. Prior approaches have tried to balance this by introducing a loss term to preserve neutral content or a regularization term to minimize changes in the model parameters, yet resolving this trade-off remains challenging. In this work, we propose to identify and preserving concepts most affected by parameter changes, termed as \textit{adversarial concepts}. This approach ensures stable erasure with minimal impact on the other concepts. We demonstrate the effectiveness of our method using the Stable Diffusion model, showing that it outperforms state-of-the-art erasure methods in eliminating unwanted content while maintaining the integrity of other unrelated elements. Our code is available at \url{https://github.com/tuananhbui89/Erasing-Adversarial-Preservation}.

Randomized controlled trials (RCTs) are the standard for evaluating the effectiveness of clinical interventions. To address the limitations of RCTs on real-world populations, we developed a methodology that uses a large observational electronic health record (EHR) dataset. Principles of regression discontinuity (rd) were used to derive randomized data subsets to test expert-driven interventions using dynamic Bayesian Networks (DBNs) do-operations. This combined method was applied to a chronic kidney disease (CKD) cohort of more than two million individuals and used to understand the associational and causal relationships of CKD variables with respect to a surrogate outcome of >=40% decline in estimated glomerular filtration rate (eGFR). The associational and causal analyses depicted similar findings across DBNs from two independent healthcare systems. The associational analysis showed that the most influential variables were eGFR, urine albumin-to-creatinine ratio, and pulse pressure, whereas the causal analysis showed eGFR as the most influential variable, followed by modifiable factors such as medications that may impact kidney function over time. This methodology demonstrates how real-world EHR data can be used to provide population-level insights to inform improved healthcare delivery.

Motion planning is an essential process for the navigation of unmanned aerial vehicles (UAVs) where they need to adapt to obstacles and different structures of their operating environment to reach the goal. This paper presents an optimal motion planner for UAVs operating in unknown complex environments. The motion planner receives point cloud data from a local range sensor and then converts it into a voxel grid representing the surrounding environment. A local trajectory guiding the UAV to the goal is then generated based on the voxel grid. This trajectory is further optimized using model predictive control (MPC) to enhance the safety, speed, and smoothness of UAV operation. The optimization is carried out via the definition of several cost functions and constraints, taking into account the UAV's dynamics and requirements. A number of simulations and comparisons with a state-of-the-art method have been conducted in a complex environment with many obstacles to evaluate the performance of our method. The results show that our method provides not only shorter and smoother trajectories but also faster and more stable speed profiles. It is also energy efficient making it suitable for various UAV applications., Comment: In proceedings of 2024, the 7th International Conference on Control, Robotics and Informatics (ICCRI 2024)

Maintaining roadway infrastructure is essential for ensuring a safe, efficient, and sustainable transportation system. However, manual data collection for detecting road damage is time-consuming, labor-intensive, and poses safety risks. Recent advancements in artificial intelligence, particularly deep learning, offer a promising solution for automating this process using road images. This paper presents a comprehensive workflow for road damage detection using deep learning models, focusing on optimizations for inference speed while preserving detection accuracy. Specifically, to accommodate hardware limitations, large images are cropped, and lightweight models are utilized. Additionally, an external pothole dataset is incorporated to enhance the detection of this underrepresented damage class. The proposed approach employs multiple model architectures, including a custom YOLOv7 model with Coordinate Attention layers and a Tiny YOLOv7 model, which are trained and combined to maximize detection performance. The models are further reparameterized to optimize inference efficiency. Experimental results demonstrate that the ensemble of the custom YOLOv7 model with three Coordinate Attention layers and the default Tiny YOLOv7 model achieves an F1 score of 0.7027 with an inference speed of 0.0547 seconds per image. The complete pipeline, including data preprocessing, model training, and inference scripts, is publicly available on the project's GitHub repository, enabling reproducibility and facilitating further research., Comment: Invited paper in the Optimized Road Damage Detection Challenge (ORDDC'2024), a track in the IEEE BigData 2024 Challenge

This research exploits the applications of reconfigurable intelligent surface (RIS)-assisted multiple input multiple output (MIMO) systems, specifically addressing the enhancement of communication reliability with modulated signals. Specifically, we first derive the analytical downlink symbol error rate (SER) of each user as a multivariate function of both the phase-shift and beamforming vectors. The analytical SER enables us to obtain insights into the synergistic dynamics between the RIS and MIMO communication. We then introduce a novel average SER minimization problem subject to the practical constraints of the transmitted power budget and phase shift coefficients, which is NP-hard. By incorporating the differential evolution (DE) algorithm as a pivotal tool for optimizing the intricate active and passive beamforming variables in RIS-assisted communication systems, the non-convexity of the considered SER optimization problem can be effectively handled. Furthermore, an efficient local search is incorporated into the DE algorithm to overcome the local optimum, and hence offer low SER and high communication reliability. Monte Carlo simulations validate the analytical results and the proposed optimization framework, indicating that the joint active and passive beamforming design is superior to the other benchmarks., Comment: 16 pages, 13 figures, and 1 table . Accepted by TWC

Whole slide image (WSI) classification is a crucial problem for cancer diagnostics in clinics and hospitals. A WSI, acquired at gigapixel size, is commonly tiled into patches and processed by multiple-instance learning (MIL) models. Previous MIL-based models designed for this problem have only been evaluated on individual tasks for specific organs, and the ability to handle multiple tasks within a single model has not been investigated. In this study, we propose MECFormer, a generative Transformer-based model designed to handle multiple tasks within one model. To leverage the power of learning multiple tasks simultaneously and to enhance the model's effectiveness in focusing on each individual task, we introduce an Expert Consultation Network, a projection layer placed at the beginning of the Transformer-based model. Additionally, to enable flexible classification, autoregressive decoding is incorporated by a language decoder for WSI classification. Through extensive experiments on five datasets involving four different organs, one cancer classification task, and four cancer subtyping tasks, MECFormer demonstrates superior performance compared to individual state-of-the-art multiple-instance learning models., Comment: Accepted for presentation at ACCV2024

Accurate and efficient theoretical techniques for describing ionic fluids are highly desirable for many applications across the physical, biological and materials sciences. With a rigorous statistical mechanical foundation, classical density functional theory (cDFT) is an appealing approach, but the competition between strong Coulombic interactions and steric repulsion limits the accuracy of current approximate functionals. Here, we extend a recently presented machine learning (ML) approach [Samm\"uller et al., Proc. Natl. Acad. Sci. USA, 120, e2312484120 (2023)] designed for systems with short-ranged interactions to ionic fluids. By adopting ideas from local molecular field theory, the framework we present amounts to using neural networks to learn the local relationship between the one-body direct correlation functions and inhomogeneous density profiles for a "mimic'' short-ranged system, with effects of long-ranged interactions accounted for in a mean-field, yet well-controlled, manner. By comparing to results from molecular simulations, we show that our approach accurately describes the structure and thermodynamics of the prototypical model for electrolyte solutions and ionic liquids: the restricted primitive model. The framework we present acts as an important step toward extending ML approaches for cDFT to systems with accurate interatomic potentials., Comment: Main: 7 pages, 3 figures. SI: 16 pages, 8 figures

Recent advancements in Code Large Language Models (CodeLLMs) have predominantly focused on open-ended code generation tasks, often neglecting the critical aspect of code understanding and comprehension. To bridge this gap, we present CodeMMLU, a comprehensive multiple-choice question-answer benchmark designed to evaluate the depth of software and code understanding in LLMs. CodeMMLU includes over 10,000 questions sourced from diverse domains, encompassing tasks such as code analysis, defect detection, and software engineering principles across multiple programming languages. Unlike traditional benchmarks, CodeMMLU assesses models's ability to reason about code rather than merely generate it, providing deeper insights into their grasp of complex software concepts and systems. Our extensive evaluation reveals that even state-of-the-art models face significant challenges with CodeMMLU, highlighting deficiencies in comprehension beyond code generation. By underscoring the crucial relationship between code understanding and effective generation, CodeMMLU serves as a vital resource for advancing AI-assisted software development, ultimately aiming to create more reliable and capable coding assistants.

Offline reinforcement learning (RL) has garnered significant attention for its ability to learn effective policies from pre-collected datasets without the need for further environmental interactions. While promising results have been demonstrated in single-agent settings, offline multi-agent reinforcement learning (MARL) presents additional challenges due to the large joint state-action space and the complexity of multi-agent behaviors. A key issue in offline RL is the distributional shift, which arises when the target policy being optimized deviates from the behavior policy that generated the data. This problem is exacerbated in MARL due to the interdependence between agents' local policies and the expansive joint state-action space. Prior approaches have primarily addressed this challenge by incorporating regularization in the space of either Q-functions or policies. In this work, we introduce a regularizer in the space of stationary distributions to better handle distributional shift. Our algorithm, ComaDICE, offers a principled framework for offline cooperative MARL by incorporating stationary distribution regularization for the global learning policy, complemented by a carefully structured multi-agent value decomposition strategy to facilitate multi-agent training. Through extensive experiments on the multi-agent MuJoCo and StarCraft II benchmarks, we demonstrate that ComaDICE achieves superior performance compared to state-of-the-art offline MARL methods across nearly all tasks.