Search

Diffractive deep neural network (D2NN), known for its high speed, low power consumption, and strong parallelism, has been widely applied across various fields, including pattern recognition, image processing, and image transmission. However, existing network architectures primarily focus on data representation within the original domain, with limited exploration of the latent space, thereby restricting the information mining capabilities and multifunctional integration of D2NNs. Here, we propose an all-optical autoencoder (OAE) framework that can encode the input wavefield into a prior shape distribution in the latent space and decode the encoded pattern back to the original wavefield. By leveraging the non-reciprocal property of D2NN, the OAE models function as encoders in one direction of wave propagation and as decoders in the opposite direction. We further apply the models to three key areas: image denoising, noise-resistant reconfigurable image classification, and image generation. Proof-of-concept experiments have been conducted to validate numerical simulations. Our OAE framework fully exploits the potential of latent space representations, enabling a single set of diffractive processors to simultaneously achieve image reconstruction, representation, and generation. It can be viewed as both a counterpart and an extension of the electronic autoencoder model. This work not only offers fresh insights into the design of optical generative models but also paves the way for developing and applying multifunctional, highly integrated, and general optical intelligent systems., Comment: 21 pages, 7 figure

Super-Resolution Ultrasound (SRUS) imaging through localising and tracking microbubbles, also known as Ultrasound Localisation Microscopy (ULM), has demonstrated significant potential for reconstructing microvasculature and flows with sub-diffraction resolution in clinical diagnostics. However, imaging organs with large tissue movements, such as those caused by respiration, presents substantial challenges. Existing methods often require breath holding to maintain accumulation accuracy, which limits data acquisition time and ULM image saturation. To improve image quality in the presence of large tissue movements, this study introduces an approach integrating high-frame-rate ultrasound with online precise robotic probe control. Tested on a microvasculature phantom with translation motions up to 20 mm, twice the aperture size of the matrix array used, our method achieved real-time tracking of the moving phantom and imaging volume rate at 85 Hz, keeping majority of the target volume in the imaging field of view. ULM images of the moving cross channels in the phantom were successfully reconstructed in post-processing, demonstrating the feasibility of super-resolution imaging under large tissue motions. This represents a significant step towards ULM imaging of organs with large motion.

Precision breast cancer (BC) risk assessment is crucial for developing individualized screening and prevention. Despite the promising potential of recent mammogram (MG) based deep learning models in predicting BC risk, they mostly overlook the 'time-to-future-event' ordering among patients and exhibit limited explorations into how they track history changes in breast tissue, thereby limiting their clinical application. In this work, we propose a novel method, named OA-BreaCR, to precisely model the ordinal relationship of the time to and between BC events while incorporating longitudinal breast tissue changes in a more explainable manner. We validate our method on public EMBED and inhouse datasets, comparing with existing BC risk prediction and time prediction methods. Our ordinal learning method OA-BreaCR outperforms existing methods in both BC risk and time-to-future-event prediction tasks. Additionally, ordinal heatmap visualizations show the model's attention over time. Our findings underscore the importance of interpretable and precise risk assessment for enhancing BC screening and prevention efforts. The code will be accessible to the public.

With the increasing integration of inverter-based resources into the power grid, there has been a notable reduction in system inertia, potentially compromising frequency stability. To assess the suitability of existing area inertia estimation techniques for real-world power systems, this paper presents a rigorous comparative analysis of system identification, measurement reconstruction, and electromechanical oscillation-based area inertia estimation methodologies, specifically applied to the large-scale and multi-area WECC 240-bus power system. Comprehensive results show that the system identification-based approach exhibits superior robustness and accuracy relative to its counterparts.

The perennial demand for highly efficient short-haul communications is evidenced by a sustained explosion of growth in data center infrastructure that is predicted to continue for the foreseeable future. In these relatively compact networks, cost-sensitivity is of particular importance, which limits options to direct detection schemes that are more cost efficient than their coherent counterparts. Since their initial demonstration, multi-soliton states in optical microresonators have been observed to manifest in self-organised ensembles where soliton pulses are equally spaced around the resonators. In the spectral domain, these states, dubbed soliton crystals (SCs), result in significant enhancements to individual comb lines depending on the crystal state, making them well suited towards intensity-modulated direct detection (IMDD) schemes. In this work, we experimentally demonstrate adiabatic, deterministic access to lower-order soliton crystal states using an auxiliary-assisted cavity pumping method, attaining up to 19.6 dB enhancement of the comb lines in the 7-SC configuration compared to the single-soliton state. Seven comb lines of each 46 Gbaud/s pulse amplitude modulation 4 (PAM4) is transmitted over 4km of fiber in comb lines across the C-band with bit-error-rates (BER) as low as 5E-5. Our demonstration shows the promising way of using soliton crystal states as future integrated sources for highly stable Terabaud/s datacenter communications.

Semiconductor nanowires have shown great potential for enabling ultra-compact lasers for integrated photonics platforms. Despite the impressive progress in developing nanowire lasers, their integration into Si photonics platforms remains challenging largely due to the use of III-V and II-VI semiconductors as gain media. These materials not only have high material costs, but also require inherently complex integration with Si-based fabrication processing, increasing overall costs and thereby limiting their large-scale adoption. Furthermore, these material-based nanowire lasers rarely emit above 2 um, which is a technologically important wavelength regime for various applications in imaging and quantum sensing. Recently, group-IV nanowires, particularly direct bandgap GeSn nanowires capable of emitting above 2 um, have emerged as promising cost-effective gain media for Si-compatible nanowire lasers, but there has been no successful demonstration of lasing from this seemingly promising nanowire platform. Herein, we report the experimental observation of lasing above 2 um from a single bottom-up grown GeSn nanowire. By harnessing strain engineering and optimized cavity designs simultaneously, the single GeSn nanowire achieves an amplified material gain that can sufficiently overcome minimized optical losses, resulting in a single-mode lasing with an ultra-low threshold of ~5.3 kW cm-2. Our finding paves the way for all-group IV mid-infrared photonic-integrated circuits with compact Si-compatible lasers for on-chip classical and quantum sensing and free-space communication., Comment: 24 pages, 4 figures

Verstraete, Dehaene, and De Moor (2003) showed that SLOCC invariants provide entanglement monotones. We observe that many highly entangled or useful four-qubit states that appear in prior literature are stationary points of such entanglement measures. This motivates the search for more stationary points. We use the notion of critical points (in the sense of the Kempf-Ness theorem) together with Vinberg theory to reduce the complexity of the problem significantly. We solve the corresponding systems utilizing modern numerical nonlinear algebra methods and reduce the solutions by natural symmetries. This method produces an extended list of four-qubit stationary points, which includes all the critical states in the survey by Enriquez et al (2016).

Humans effortlessly integrate common-sense knowledge with sensory input from vision and touch to understand their surroundings. Emulating this capability, we introduce FusionSense, a novel 3D reconstruction framework that enables robots to fuse priors from foundation models with highly sparse observations from vision and tactile sensors. FusionSense addresses three key challenges: (i) How can robots efficiently acquire robust global shape information about the surrounding scene and objects? (ii) How can robots strategically select touch points on the object using geometric and common-sense priors? (iii) How can partial observations such as tactile signals improve the overall representation of the object? Our framework employs 3D Gaussian Splatting as a core representation and incorporates a hierarchical optimization strategy involving global structure construction, object visual hull pruning and local geometric constraints. This advancement results in fast and robust perception in environments with traditionally challenging objects that are transparent, reflective, or dark, enabling more downstream manipulation or navigation tasks. Experiments on real-world data suggest that our framework outperforms previously state-of-the-art sparse-view methods. All code and data are open-sourced on the project website.

Discrete diffusion models have achieved success in tasks like image generation and masked language modeling but face limitations in controlled content editing. We introduce DICE (Discrete Inversion for Controllable Editing), the first approach to enable precise inversion for discrete diffusion models, including multinomial diffusion and masked generative models. By recording noise sequences and masking patterns during the reverse diffusion process, DICE enables accurate reconstruction and flexible editing of discrete data without the need for predefined masks or attention manipulation. We demonstrate the effectiveness of DICE across both image and text domains, evaluating it on models such as VQ-Diffusion, Paella, and RoBERTa. Our results show that DICE preserves high data fidelity while enhancing editing capabilities, offering new opportunities for fine-grained content manipulation in discrete spaces. For project webpage, see https://hexiaoxiao-cs.github.io/DICE/.

Current frontier video diffusion models have demonstrated remarkable results at generating high-quality videos. However, they can only generate short video clips, normally around 10 seconds or 240 frames, due to computation limitations during training. In this work, we show that existing models can be naturally extended to autoregressive video diffusion models without changing the architectures. Our key idea is to assign the latent frames with progressively increasing noise levels rather than a single noise level, which allows for fine-grained condition among the latents and large overlaps between the attention windows. Such progressive video denoising allows our models to autoregressively generate video frames without quality degradation or abrupt scene changes. We present state-of-the-art results on long video generation at 1 minute (1440 frames at 24 FPS). Videos from this paper are available at https://desaixie.github.io/pa-vdm/., Comment: 15 pages, 5 figures. Our video results and code are available at https://desaixie.github.io/pa-vdm/

Large Language Models (LLMs) exhibit impressive performance across various domains but still struggle with arithmetic reasoning tasks. Recent work shows the effectiveness of prompt design methods in enhancing reasoning capabilities. However, these approaches overlook crucial requirements for prior knowledge of specific concepts, theorems, and tricks to tackle most arithmetic reasoning problems successfully. To address this issue, we propose a novel and effective Teaching-Inspired Integrated Framework, which emulates the instructional process of a teacher guiding students. This method equips LLMs with essential concepts, relevant theorems, and similar problems with analogous solution approaches, facilitating the enhancement of reasoning abilities. Additionally, we introduce two new Chinese datasets, MathMC and MathToF, both with detailed explanations and answers. Experiments are conducted on nine benchmarks which demonstrates that our approach improves the reasoning accuracy of LLMs. With GPT-4 and our framework, we achieve new state-of-the-art performance on four math benchmarks (AddSub, SVAMP, Math23K and AQuA) with accuracies of 98.2% (+3.3%), 93.9% (+0.2%), 94.3% (+7.2%) and 81.1% (+1.2%). Our data and code are available at https://github.com/SallyTan13/Teaching-Inspired-Prompting.

Providing feedback is widely recognized as crucial for refining students' writing skills. Recent advances in language models (LMs) have made it possible to automatically generate feedback that is actionable and well-aligned with human-specified attributes. However, it remains unclear whether the feedback generated by these models is truly effective in enhancing the quality of student revisions. Moreover, prompting LMs with a precise set of instructions to generate feedback is nontrivial due to the lack of consensus regarding the specific attributes that can lead to improved revising performance. To address these challenges, we propose PROF that PROduces Feedback via learning from LM simulated student revisions. PROF aims to iteratively optimize the feedback generator by directly maximizing the effectiveness of students' overall revising performance as simulated by LMs. Focusing on an economic essay assignment, we empirically test the efficacy of PROF and observe that our approach not only surpasses a variety of baseline methods in effectiveness of improving students' writing but also demonstrates enhanced pedagogical values, even though it was not explicitly trained for this aspect., Comment: Accepted to EMNLP 2024

Kolmogorov--Arnold Networks (KANs), a recently proposed neural network architecture, have gained significant attention in the deep learning community, due to their potential as a viable alternative to multi-layer perceptrons (MLPs) and their broad applicability to various scientific tasks. Empirical investigations demonstrate that KANs optimized via stochastic gradient descent (SGD) are capable of achieving near-zero training loss in various machine learning (e.g., regression, classification, and time series forecasting, etc.) and scientific tasks (e.g., solving partial differential equations). In this paper, we provide a theoretical explanation for the empirical success by conducting a rigorous convergence analysis of gradient descent (GD) and SGD for two-layer KANs in solving both regression and physics-informed tasks. For regression problems, we establish using the neural tangent kernel perspective that GD achieves global linear convergence of the objective function when the hidden dimension of KANs is sufficiently large. We further extend these results to SGD, demonstrating a similar global convergence in expectation. Additionally, we analyze the global convergence of GD and SGD for physics-informed KANs, which unveils additional challenges due to the more complex loss structure. This is the first work establishing the global convergence guarantees for GD and SGD applied to optimize KANs and physics-informed KANs.

As the quantum research community continues to grow and new algorithms are designed, developed, and implemented, it is crucial to start thinking about security aspects and potential threats that could result in misuse of the algorithms, or jeopardize the information processed with these quantum algorithms. This work focuses on exploration of two types of potential attacks that could be deployed on a cloud-based quantum computer by an attacker circuit trying to interfere with victim circuit. The two attacks, called Improper Initialization Attack (IIA) and Higher Energy Attack (HEA), are for the first time applied to a well-known and widely used quantum algorithm: HHL. The HHL algorithm is used in the field of machine learning and big data for solving systems of linear equations. This work evaluates the effect of the attacks on different qubits within the HHL algorithm: ancilla qubit, clock qubit, and b qubit. This work demonstrates that the two attacks are able to cause incorrect results, even when only one of the qubits in the victim algorithm is attacked. Having discovered the vulnerabilities, the work motivates the need for future work to develop defense strategies for each of these attack scenarios.

Bimanual manipulation is essential in robotics, yet developing foundation models is extremely challenging due to the inherent complexity of coordinating two robot arms (leading to multi-modal action distributions) and the scarcity of training data. In this paper, we present the Robotics Diffusion Transformer (RDT), a pioneering diffusion foundation model for bimanual manipulation. RDT builds on diffusion models to effectively represent multi-modality, with innovative designs of a scalable Transformer to deal with the heterogeneity of multi-modal inputs and to capture the nonlinearity and high frequency of robotic data. To address data scarcity, we further introduce a Physically Interpretable Unified Action Space, which can unify the action representations of various robots while preserving the physical meanings of original actions, facilitating learning transferrable physical knowledge. With these designs, we managed to pre-train RDT on the largest collection of multi-robot datasets to date and scaled it up to 1.2B parameters, which is the largest diffusion-based foundation model for robotic manipulation. We finally fine-tuned RDT on a self-created multi-task bimanual dataset with over 6K+ episodes to refine its manipulation capabilities. Experiments on real robots demonstrate that RDT significantly outperforms existing methods. It exhibits zero-shot generalization to unseen objects and scenes, understands and follows language instructions, learns new skills with just 1~5 demonstrations, and effectively handles complex, dexterous tasks. We refer to https://rdt-robotics.github.io/rdt-robotics/ for the code and videos., Comment: 10 pages, conference

BCC lattices with twinned meta-crystal architecture inspired by strengthening of bulk metals have significantly improved mechanical performance; however, their deformation behaviour and underlying strengthening mechanisms remain unclear. Here, we reveal that twinning causes a transition from bending to stretch-dominated behaviour in BCC lattices, violating the Gibson-Ashby model, and eliciting vast improvements in stiffness (+181%) and strength (+128%). By controlling a heterogenous distribution of twinned grain boundaries, inspired by bimodal harmonic microstructure, we amplify the axial strain energy at location specific sites, further enhancing the stiffness of twinned BCC lattices by 11.1%. Our lattice design philosophy unleashes the potential of cellular materials for high-performance engineering applications.

Large Language Models (LLMs) have demonstrated exceptional promise in translation tasks for high-resource languages. However, their performance in low-resource languages is limited by the scarcity of both parallel and monolingual corpora, as well as the presence of noise. Consequently, such LLMs suffer with alignment and have lagged behind State-of-The-Art (SoTA) neural machine translation (NMT) models in these settings. This paper introduces NusaMT-7B, an LLM-based machine translation model for low-resource Indonesian languages, starting with Balinese and Minangkabau. Leveraging the pretrained LLaMA2-7B, our approach integrates continued pre-training on monolingual data, Supervised Fine-Tuning (SFT), self-learning, and an LLM-based data cleaner to reduce noise in parallel sentences. In the FLORES-200 multilingual translation benchmark, NusaMT-7B outperforms SoTA models in the spBLEU metric by up to +6.69 spBLEU in translations into Balinese and Minangkabau, but underperforms by up to -3.38 spBLEU in translations into higher-resource languages. Our results show that fine-tuned LLMs can enhance translation quality for low-resource languages, aiding in linguistic preservation and cross-cultural communication., Comment: Accepted to SoLaR @ NeurIPS 2024

The construction of Vectorized High-Definition (HD) map typically requires capturing both category and geometry information of map elements. Current state-of-the-art methods often adopt solely either point-level or instance-level representation, overlooking the strong intrinsic relationships between points and instances. In this work, we propose a simple yet efficient framework named MGMapNet (Multi-Granularity Map Network) to model map element with a multi-granularity representation, integrating both coarse-grained instance-level and fine-grained point-level queries. Specifically, these two granularities of queries are generated from the multi-scale bird's eye view (BEV) features using a proposed Multi-Granularity Aggregator. In this module, instance-level query aggregates features over the entire scope covered by an instance, and the point-level query aggregates features locally. Furthermore, a Point Instance Interaction module is designed to encourage information exchange between instance-level and point-level queries. Experimental results demonstrate that the proposed MGMapNet achieves state-of-the-art performance, surpassing MapTRv2 by 5.3 mAP on nuScenes and 4.4 mAP on Argoverse2 respectively.

We propose a {\em novel} piecewise stationary linear bandit (PSLB) model, where the environment randomly samples a context from an unknown probability distribution at each changepoint, and the quality of an arm is measured by its return averaged over all contexts. The contexts and their distribution, as well as the changepoints are unknown to the agent. We design {\em Piecewise-Stationary $\varepsilon$-Best Arm Identification$^+$} (PS$\varepsilon$BAI$^+$), an algorithm that is guaranteed to identify an $\varepsilon$-optimal arm with probability $\ge 1-\delta$ and with a minimal number of samples. PS$\varepsilon$BAI$^+$ consists of two subroutines, PS$\varepsilon$BAI and {\sc Na\"ive $\varepsilon$-BAI} (N$\varepsilon$BAI), which are executed in parallel. PS$\varepsilon$BAI actively detects changepoints and aligns contexts to facilitate the arm identification process. When PS$\varepsilon$BAI and N$\varepsilon$BAI are utilized judiciously in parallel, PS$\varepsilon$BAI$^+$ is shown to have a finite expected sample complexity. By proving a lower bound, we show the expected sample complexity of PS$\varepsilon$BAI$^+$ is optimal up to a logarithmic factor. We compare PS$\varepsilon$BAI$^+$ to baseline algorithms using numerical experiments which demonstrate its efficiency. Both our analytical and numerical results corroborate that the efficacy of PS$\varepsilon$BAI$^+$ is due to the delicate change detection and context alignment procedures embedded in PS$\varepsilon$BAI., Comment: 69 pages. Accepted to NeurIPS 2024

Current methods of building LLMs with voice interaction capabilities rely heavily on explicit text autoregressive generation before or during speech response generation to maintain content quality, which unfortunately brings computational overhead and increases latency in multi-turn interactions. To address this, we introduce IntrinsicVoic,e an LLM designed with intrinsic real-time voice interaction capabilities. IntrinsicVoice aims to facilitate the transfer of textual capabilities of pre-trained LLMs to the speech modality by mitigating the modality gap between text and speech. Our novelty architecture, GroupFormer, can reduce speech sequences to lengths comparable to text sequences while generating high-quality audio, significantly reducing the length difference between speech and text, speeding up inference, and alleviating long-text modeling issues. Additionally, we construct a multi-turn speech-to-speech dialogue dataset named \method-500k which includes nearly 500k turns of speech-to-speech dialogues, and a cross-modality training strategy to enhance the semantic alignment between speech and text. Experimental results demonstrate that IntrinsicVoice can generate high-quality speech response with latency lower than 100ms in multi-turn dialogue scenarios. Demos are available at https://instrinsicvoice.github.io/.

The Mpemba effect is a counterintuitive physical phenomenon where a hot system cools faster than a warm one. In recent years, theoretical analyses of the Mpemba effect have been developed for microscopic systems and experimentally verified. However, the conventional theory relies on a specific choice of distance measure to quantify relaxation speed, leading to several theoretical ambiguities. In this Letter, we derive a rigorous quantification of the Mpemba effect based on thermomajorization theory, referred to as the thermomajorization Mpemba effect. This approach resolves all existing ambiguities and provides a unification of the conventional Mpemba effect across all monotone measures. Furthermore, we demonstrate the universality of the thermomajorization Mpemba effect for Markovian dynamics, rigorously proving that it can occur in any temperature regime with fixed energy levels. Our findings offer a paradigm shift in understanding thermal relaxation processes and provide insights into designing systems that exhibit the Mpemba effect., Comment: 7+6 pages, 3+0 figures

This paper studies an online selection problem, where a seller seeks to sequentially sell multiple copies of an item to arriving buyers. We consider an adversarial setting, making no modeling assumptions about buyers' valuations for the items except acknowledging a finite support. In this paper, we focus on a class of static pricing algorithms that sample a price from a pre-determined distribution and sell items to buyers whose valuations exceed the sampled price. Such algorithms are of practical interests due to their advantageous properties, such as ease of implementation and non-discrimination over prices. Our work shows that the simple static pricing strategy can achieve strong guarantees comparable to the best known dynamic pricing algorithms. Particularly, we design the optimal static pricing algorithms for the adversarial online selection problem and its two important variants: the online assignment problem and the online selection with convex cost. The static pricing algorithms can even attain the optimal competitive ratios among all online algorithms for the online selection problem and the online assignment problem. To achieve these results, we propose an economics-based approach in the competitive analysis of static pricing algorithms, and develop a novel representative function-based approach to derive the lower bounds. We expect these approaches will be useful in related problems such as online matching.

Mediation is a dispute resolution method featuring a neutral third-party (mediator) who intervenes to help the individuals resolve their dispute. In this paper, we investigate to which extent large language models (LLMs) are able to act as mediators. We investigate whether LLMs are able to analyze dispute conversations, select suitable intervention types, and generate appropriate intervention messages. Using a novel, manually created dataset of 50 dispute scenarios, we conduct a blind evaluation comparing LLMs with human annotators across several key metrics. Overall, the LLMs showed strong performance, even outperforming our human annotators across dimensions. Specifically, in 62% of the cases, the LLMs chose intervention types that were rated as better than or equivalent to those chosen by humans. Moreover, in 84% of the cases, the intervention messages generated by the LLMs were rated as better than or equal to the intervention messages written by humans. LLMs likewise performed favourably on metrics such as impartiality, understanding and contextualization. Our results demonstrate the potential of integrating AI in online dispute resolution (ODR) platforms.

Accurately measuring magnetic field strength in the interstellar medium, including giant molecular clouds (GMCs), remains a significant challenge. We present a machine learning approach using Denoising Diffusion Probabilistic Models (DDPMs) to estimate magnetic field strength from synthetic observables such as column density, dust continuum polarization vector orientation angles, and line-of-sight (LOS) nonthermal velocity dispersion. We trained three versions of the DDPM model: the 1-channel DDPM (using only column density), the 2-channel DDPM (incorporating both column density and polarization angles), and the 3-channel DDPM (which combines column density, polarization angles, and LOS nonthermal velocity dispersion). We assessed the models on both synthetic test samples and new simulation data that were outside the training set's distribution. The 3-channel DDPM consistently outperformed both the other DDPM variants and the power-law fitting approach based on column density alone, demonstrating its robustness in handling previously unseen data. Additionally, we compared the performance of the Davis-Chandrasekhar-Fermi (DCF) methods, both classical and modified, to the DDPM predictions. The classical DCF method overestimated the magnetic field strength by approximately an order of magnitude. Although the modified DCF method showed improvement over the classical version, it still fell short of the precision achieved by the 3-channel DDPM., Comment: submitted to ApJ, comments welcome

In this work, we study the gravitational quasinormal modes and the gravitational echoes of a double braneworld. The double braneworld is a kind of split thick brane, which is crucial for addressing the hierarchy problem in the thick brane scenarios. Using the Bernstein spectral method, direct integration method, and asymptotic iteration method, we calculate the quasinormal mode frequencies of the double brane. We find that the quasinormal spectrum is very different from that of the single brane model, especially the high overtone mode. We also perform numerical evolution to study the time-domain properties of the characteristic modes of the double brane. The results show that when the degree of brane splitting is large, gravitational echoes of oscillation attenuation between sub-branes will appear in the thick brane. Furthermore, different long-lived Kaluza-Klein modes interfere with each other, resulting in a beating effect. Compared to a single brane model, the phenomenon of the split double brane is richer, and the lifetime of the massive Kaluza-Klein graviton of the double brane is longer. These phenomena may have potential phenomenological interest. We hope to detect these extra-dimensional signals in future gravitational wave detectors or accelerators., Comment: 14 pages, 16 figures

We propose a fourth-order cut-cell method for solving Poisson's equations in three-dimensional irregular domains. Major distinguishing features of our method include (a) applicable to arbitrarily complex geometries, (b) high order discretization, (c) optimal complexity. Feature (a) is achieved by Yin space, which is a mathematical model for three-dimensional continua. Feature (b) is accomplished by poised lattice generation (PLG) algorithm, which finds stencils near the irregular boundary for polynomial fitting. Besides, for feature (c), we design a modified multigrid solver whose complexity is theoretically optimal by applying nested dissection (ND) ordering method.

We study EgalMAB, an egalitarian assignment problem in the context of stochastic multi-armed bandits. In EgalMAB, an agent is tasked with assigning a set of users to arms. At each time step, the agent must assign exactly one arm to each user such that no two users are assigned to the same arm. Subsequently, each user obtains a reward drawn from the unknown reward distribution associated with its assigned arm. The agent's objective is to maximize the minimum expected cumulative reward among all users over a fixed horizon. This problem has applications in areas such as fairness in job and resource allocations, among others. We design and analyze a UCB-based policy EgalUCB and establish upper bounds on the cumulative regret. In complement, we establish an almost-matching policy-independent impossibility result.

Two dimensional (2D) crystals have attracted strong interest as a new class of proton conducting materials that can block atoms, molecules and ions while allowing proton transport through the atomically thin basal planes. Although 2D materials exhibit this perfect selectivity, the reported proton conductivities have been relatively low. Here we show that vacancy-rich titania monolayers are highly permeable to protons while remaining impermeable to helium with proton conductivity exceeding 100 S cm-2 at 200 C and surpassing targets set by industry roadmaps. The fast and selective proton transport is attributed to an extremely high density of titanium-atom vacancies (one per square nm), which effectively turns titania monolayers into angstrom-scale sieves. Our findings highlight the potential of 2D oxides as membrane materials for hydrogen-based technologies.

This study focuses on a critical aspect of bridge engineering -- the evaluation of cable forces, paying particular attention to the cables that are internally constrained by elastic supports. Detecting these cable forces is important for the safety and stability of bridges. The practical problem introduces a novel mathematical challenge: how to effectively address string vibration equations with one or multiple internal elastic supports,~which remains a theoretical issue not fully solved in engineering. To tackle this, it is necessary to firstly establish an appropriate mathematical model and accurately define initial-boundary value problems. We then formulate the well-posedness of the solution using both classical and weak solution approaches, supplementing the existing numerical results available in engineering. Meanwhile, we attempt to use PINNs (Physics-Informed Neural Networks) instead of traditional FEM (Finite Element Method) in engineering. Consequently, in contrast to the classical solution method, we demonstrate that for a string with finite elastic supports, the weak solution method not only improves mathematical modeling efficiency but also simplifies the process of explaining the well-posedness of the solution.

Recent advancements in prompt engineering strategies, such as Chain-of-Thought (CoT) and Self-Discover, have demonstrated significant potential in improving the reasoning abilities of Large Language Models (LLMs). However, these state-of-the-art (SOTA) prompting strategies rely on single or fixed set of static seed reasoning modules like \emph{"think step by step"} or \emph{"break down this problem"} intended to simulate human approach to problem-solving. This constraint limits the flexibility of models in tackling diverse problems effectively. In this paper, we introduce Auto-Evolve, a novel framework that enables LLMs to self-create dynamic reasoning modules and downstream action plan, resulting in significant improvements over current SOTA methods. We evaluate Auto-Evolve on the challenging BigBench-Hard (BBH) dataset with Claude 2.0, Claude 3 Sonnet, Mistral Large, and GPT 4, where it consistently outperforms the SOTA prompt strategies. Auto-Evolve outperforms CoT by up to 10.4\% and on an average by 7\% across these four models. Our framework introduces two innovations: a) Auto-Evolve dynamically generates reasoning modules for each task while aligning with human reasoning paradigm, thus eliminating the need for predefined templates. b) We introduce an iterative refinement component, that incrementally refines instruction guidance for LLMs and helps boost performance by average 2.8\% compared to doing it in a single step., Comment: Accepted at EMNLP 2024

Mixture-of-Experts large language models (MoE-LLMs) marks a significant step forward of language models, however, they encounter two critical challenges in practice: 1) expert parameters lead to considerable memory consumption and loading latency; and 2) the current activated experts are redundant, as many tokens may only require a single expert. Motivated by these issues, we investigate the MoE-LLMs and make two key observations: a) different experts exhibit varying behaviors on activation reconstruction error, routing scores, and activated frequencies, highlighting their differing importance, and b) not all tokens are equally important -- only a small subset is critical. Building on these insights, we propose MC-MoE, a training-free Mixture-Compressor for MoE-LLMs, which leverages the significance of both experts and tokens to achieve an extreme compression. First, to mitigate storage and loading overheads, we introduce Pre-Loading Mixed-Precision Quantization, which formulates the adaptive bit-width allocation as a Linear Programming problem, where the objective function balances multi-factors reflecting the importance of each expert. Additionally, we develop Online Dynamic Pruning, which identifies important tokens to retain and dynamically select activated experts for other tokens during inference to optimize efficiency while maintaining performance. Our MC-MoE integrates static quantization and dynamic pruning to collaboratively achieve extreme compression for MoE-LLMs with less accuracy loss, ensuring an optimal trade-off between performance and efficiency. Extensive experiments confirm the effectiveness of our approach. For instance, at 2.54 bits, MC-MoE compresses 76.6% of the model, with only a 3.8% average accuracy loss. During dynamic inference, we further reduce activated parameters by 15%, with a performance drop of less than 0.6%., Comment: 18 pages

We propose RelitLRM, a Large Reconstruction Model (LRM) for generating high-quality Gaussian splatting representations of 3D objects under novel illuminations from sparse (4-8) posed images captured under unknown static lighting. Unlike prior inverse rendering methods requiring dense captures and slow optimization, often causing artifacts like incorrect highlights or shadow baking, RelitLRM adopts a feed-forward transformer-based model with a novel combination of a geometry reconstructor and a relightable appearance generator based on diffusion. The model is trained end-to-end on synthetic multi-view renderings of objects under varying known illuminations. This architecture design enables to effectively decompose geometry and appearance, resolve the ambiguity between material and lighting, and capture the multi-modal distribution of shadows and specularity in the relit appearance. We show our sparse-view feed-forward RelitLRM offers competitive relighting results to state-of-the-art dense-view optimization-based baselines while being significantly faster. Our project page is available at: https://relit-lrm.github.io/., Comment: webpage: https://relit-lrm.github.io/

In the social world, humans possess the capability to infer and reason about others mental states (such as emotions, beliefs, and intentions), known as the Theory of Mind (ToM). Simultaneously, humans own mental states evolve in response to social situations, a capability we refer to as socialization. Together, these capabilities form the foundation of human social interaction. In the era of artificial intelligence (AI), especially with the development of large language models (LLMs), we raise an intriguing question: How do LLMs perform in terms of ToM and socialization capabilities? And more broadly, can these AI models truly enter and navigate the real social world? Existing research evaluating LLMs ToM and socialization capabilities by positioning LLMs as passive observers from a third person perspective, rather than as active participants. However, compared to the third-person perspective, observing and understanding the world from an egocentric first person perspective is a natural approach for both humans and AI agents. The ToM and socialization capabilities of LLMs from a first person perspective, a crucial attribute for advancing embodied AI agents, remain unexplored. To answer the aforementioned questions and bridge the research gap, we introduce EgoSocialArena, a novel framework designed to evaluate and investigate the ToM and socialization capabilities of LLMs from a first person perspective. It encompasses two evaluation environments: static environment and interactive environment, with seven scenarios: Daily Life, Counterfactual, New World, Blackjack, Number Guessing, and Limit Texas Hold em, totaling 2,195 data entries. With EgoSocialArena, we have conducted a comprehensive evaluation of nine advanced LLMs and observed some key insights regarding the future development of LLMs as well as the capabilities levels of the most advanced LLMs currently available., Comment: 15 pages, 5 figures

High-speed ultra-broadband detectors play a crucial role in aerospace technology, and national security etc. The interfacial work function internal photoemission (IWIP) detector employs multiple absorption mechanism comprehensively across different wavelength band to achieve complete photon type detection, which makes it possible to realize high-speed and ultra-broadband simultaneously. We propose a ratchet heterojunction IWIP (HEIWIP) detector, which shows 3-165THz ultra-broadband coverage. The high-speed response is investigated in detail by both microwave rectification technology and high-speed modulated terahertz light. Up to 5.1GHz 3dB bandwidth is acquired in terms of microwave rectification measurement. And 4.255GHz inter-mode optical beat note signal was successfully detected.

Magnetic kagome materials provide a fascinating playground for exploring the interplay of magnetism, correlation and topology. Many magnetic kagome systems have been reported including the binary FemXn (X=Sn, Ge; m:n = 3:1, 3:2, 1:1) family and the rare earth RMn6Sn6 (R = rare earth) family, where their kagome flat bands are calculated to be near the Fermi level in the paramagnetic phase. While partially filling a kagome flat band is predicted to give rise to a Stoner-type ferromagnetism, experimental visualization of the magnetic splitting across the ordering temperature has not been reported for any of these systems due to the high ordering temperatures, hence leaving the nature of magnetism in kagome magnets an open question. Here, we probe the electronic structure with angle-resolved photoemission spectroscopy in a kagome magnet thin film FeSn synthesized using molecular beam epitaxy. We identify the exchange-split kagome flat bands, whose splitting persists above the magnetic ordering temperature, indicative of a local moment picture. Such local moments in the presence of the topological flat band are consistent with the compact molecular orbitals predicted in theory. We further observe a large spin-orbital selective band renormalization in the Fe d_xy+d_(x^2-y^2 ) spin majority channel reminiscent of the orbital selective correlation effects in the iron-based superconductors. Our discovery of the coexistence of local moments with topological flat bands in a kagome system echoes similar findings in magic-angle twisted bilayer graphene, and provides a basis for theoretical effort towards modeling correlation effects in magnetic flat band systems.

For many binary mixtures, the three-phase solid-liquid-vapor equilibrium curve has intermediate pressures that are higher than the pressure at the two pure triple points. This curve shape results in a negative slope in the high-temperature region near the triple point of the less volatile component. When freezing mixtures in the negative slope regime, fluid trapped below confined ice has latent heat released with more vapor upon cooling, and thus increases in pressure. If the rising pressure of the confined fluid overcomes the strength of the confining solid, which may be its own ice, it can produce an abrupt outburst of material and an increase in the system's overall pressure. Here, we report experimental results of freezing-induced outbursts occurring in the N2/CH4, CO/CH4, and N2/C2H6 systems, and provide insight into the phenomenon through a thermodynamics perspective. We also propose other binary systems that may experience outbursts and explore the geological implications for icy worlds like Titan, Triton, Pluto and Eris, as well as rocky bodies, specifically Earth and Mars., Comment: 30 pages, 8 figures

Speech enhancement is critical for improving speech intelligibility and quality in various audio devices. In recent years, deep learning-based methods have significantly improved speech enhancement performance, but they often come with a high computational cost, which is prohibitive for a large number of edge devices, such as headsets and hearing aids. This work proposes an ultra-low-power speech enhancement system based on the brain-inspired spiking neural network (SNN) called Spiking-FullSubNet. Spiking-FullSubNet follows a full-band and sub-band fusioned approach to effectively capture both global and local spectral information. To enhance the efficiency of computationally expensive sub-band modeling, we introduce a frequency partitioning method inspired by the sensitivity profile of the human peripheral auditory system. Furthermore, we introduce a novel spiking neuron model that can dynamically control the input information integration and forgetting, enhancing the multi-scale temporal processing capability of SNN, which is critical for speech denoising. Experiments conducted on the recent Intel Neuromorphic Deep Noise Suppression (N-DNS) Challenge dataset show that the Spiking-FullSubNet surpasses state-of-the-art methods by large margins in terms of both speech quality and energy efficiency metrics. Notably, our system won the championship of the Intel N-DNS Challenge (Algorithmic Track), opening up a myriad of opportunities for ultra-low-power speech enhancement at the edge. Our source code and model checkpoints are publicly available at https://github.com/haoxiangsnr/spiking-fullsubnet., Comment: under review

While often omitted in the SMEFT analyses of electroweak measurements, the electroweak dipole operators of the bottom quark have been found to be important in some cases, and are also related to processes involving the top quark. In this paper, we further investigate their effects, focusing on the measurements of the $e^+e^-\to b\bar{b}$ process at a future lepton collider. Their linear contributions are suppressed by the bottom mass due to the helicity flip in the interference term with the SM amplitude, leading to a nontrivial interplay between the linear and the quadratic contributions. With two runs at the Z-pole and 240 GeV, the effects of CP-even dipole coefficients can be well separated from the modifications of the SM $Zb\bar{b}$ couplings, while an additional run (e.g. at 360 GeV) is useful for lifting a nontrivial second best-fit point due to the quadratic contributions. The dipole operators also offer an alternative explanation to the long-standing $A^{0,b}_{\rm FB}$ discrepancy at LEP, which can be tested at a future lepton collider., Comment: 22 pages including references, 20 figures

Text-to-video (T2V) models like Sora have made significant strides in visualizing complex prompts, which is increasingly viewed as a promising path towards constructing the universal world simulator. Cognitive psychologists believe that the foundation for achieving this goal is the ability to understand intuitive physics. However, the capacity of these models to accurately represent intuitive physics remains largely unexplored. To bridge this gap, we introduce PhyGenBench, a comprehensive \textbf{Phy}sics \textbf{Gen}eration \textbf{Ben}chmark designed to evaluate physical commonsense correctness in T2V generation. PhyGenBench comprises 160 carefully crafted prompts across 27 distinct physical laws, spanning four fundamental domains, which could comprehensively assesses models' understanding of physical commonsense. Alongside PhyGenBench, we propose a novel evaluation framework called PhyGenEval. This framework employs a hierarchical evaluation structure utilizing appropriate advanced vision-language models and large language models to assess physical commonsense. Through PhyGenBench and PhyGenEval, we can conduct large-scale automated assessments of T2V models' understanding of physical commonsense, which align closely with human feedback. Our evaluation results and in-depth analysis demonstrate that current models struggle to generate videos that comply with physical commonsense. Moreover, simply scaling up models or employing prompt engineering techniques is insufficient to fully address the challenges presented by PhyGenBench (e.g., dynamic scenarios). We hope this study will inspire the community to prioritize the learning of physical commonsense in these models beyond entertainment applications. We will release the data and codes at https://github.com/OpenGVLab/PhyGenBench, Comment: Project Page: https://phygenbench123.github.io/

As Large Language Models (LLMs) excel across tasks and specialized domains, scaling LLMs based on existing models has garnered significant attention, which faces the challenge of decreasing performance when combining disparate models. Various techniques have been proposed for the aggregation of pre-trained LLMs, including model merging, Mixture-of-Experts, and stacking. Despite their merits, a comprehensive comparison and synergistic application of them to a diverse model zoo is yet to be adequately addressed. In light of this research gap, this paper introduces Model-GLUE, a holistic LLM scaling guideline. First, our work starts with a benchmarking of existing LLM scaling techniques, especially selective merging, and variants of mixture. Utilizing the insights from the benchmark results, we formulate an strategy for the selection and aggregation of a heterogeneous model zoo characterizing different architectures and initialization. Our methodology involves the clustering of mergeable models and optimal merging strategy selection, and the integration of clusters through a model mixture. Finally, evidenced by our experiments on a diverse Llama-2-based model zoo, Model-GLUE shows an average performance enhancement of 5.61%, achieved without additional training. Codes are available at: https://github.com/Model-GLUE/Model-GLUE., Comment: 24 pages, 4 figures, accepted to NeurIPS 2024 Datasets and Benchmarks Track

We present a novel approach to classify causal micro-narratives from text. These narratives are sentence-level explanations of the cause(s) and/or effect(s) of a target subject. The approach requires only a subject-specific ontology of causes and effects, and we demonstrate it with an application to inflation narratives. Using a human-annotated dataset spanning historical and contemporary US news articles for training, we evaluate several large language models (LLMs) on this multi-label classification task. The best-performing model--a fine-tuned Llama 3.1 8B--achieves F1 scores of 0.87 on narrative detection and 0.71 on narrative classification. Comprehensive error analysis reveals challenges arising from linguistic ambiguity and highlights how model errors often mirror human annotator disagreements. This research establishes a framework for extracting causal micro-narratives from real-world data, with wide-ranging applications to social science research., Comment: Accepted to EMNLP 2024 Workshop on Narrative Understanding

The self-consistent inclusion of plasma effects in opacity calculations is a significant modeling challenge. As density increases, such effects can no longer be treated perturbatively. Building on a recently published model that addresses this challenge, we calculate opacities of oxygen at solar interior conditions. The new model includes the effects of treating the free electrons consistently with the bound electrons, and the influence of free electron energy and entropy variations are explored. It is found that, relative to a state-of-the-art-model that does not include these effects, the bound free-opacity of the oxygen plasmas considered can increase by 10%.

Designing neural network architectures that can handle data symmetry is crucial. This is especially important for geometric graphs whose properties are equivariance under Euclidean transformations. Current equivariant graph neural networks (EGNNs), particularly those using message passing, have a limitation in expressive power. Recent high-order graph neural networks can overcome this limitation, yet they lack equivariance properties, representing a notable drawback in certain applications in chemistry and physical sciences. In this paper, we introduce the Clifford Group Equivariant Graph Neural Networks (CG-EGNNs), a novel EGNN that enhances high-order message passing by integrating high-order local structures in the context of Clifford algebras. As a key benefit of using Clifford algebras, CG-EGNN can learn functions that capture equivariance from positional features. By adopting the high-order message passing mechanism, CG-EGNN gains richer information from neighbors, thus improving model performance. Furthermore, we establish the universality property of the $k$-hop message passing framework, showcasing greater expressive power of CG-EGNNs with additional $k$-hop message passing mechanism. We empirically validate that CG-EGNNs outperform previous methods on various benchmarks including n-body, CMU motion capture, and MD17, highlighting their effectiveness in geometric deep learning.

Human fact-checkers have specialized domain knowledge that allows them to formulate precise questions to verify information accuracy. However, this expert-driven approach is labor-intensive and is not scalable, especially when dealing with complex multimodal misinformation. In this paper, we propose a fully-automated framework, LRQ-Fact, for multimodal fact-checking. Firstly, the framework leverages Vision-Language Models (VLMs) and Large Language Models (LLMs) to generate comprehensive questions and answers for probing multimodal content. Next, a rule-based decision-maker module evaluates both the original content and the generated questions and answers to assess the overall veracity. Extensive experiments on two benchmarks show that LRQ-Fact improves detection accuracy for multimodal misinformation. Moreover, we evaluate its generalizability across different model backbones, offering valuable insights for further refinement.