Search

Both linear and nonlinear self-accelerating valley Hall edge states are predicted in the composited inversion-symmetry-broken photonic graphene lattice with a domain wall. The linear one that is obtained by superimposing a finite-energy Airy envelope to the valley Hall edge state shows self-accelerating, non-diffracting and self-healing properties, with the accelerating trajectory not exactly a parabola. For the nonlinear one which exhibits well self-accelerating and non-diffracting properties that may turn over the moving direction of the valley Hall edge state, it is constructed by superimposing the self-trapped self-accelerating envelope that is obtained from the envelope equation to the valley Hall edge state. The nonlinear self-accelerating valley Hall edge state can circumvent sharp corners without back-scattering, but it is impossible to completely prohibit radiating into the bulk because the topological protection and the self-accelerating demand contrary length of the oscillating tail. The results exhibit the possibility on manipulating topological edge states via self-accelerating waves, and pave the way for the self-accelerating topological photonics., Comment: 7 pages, 4 figures; comments are welcome

The exploration of various vision-language tasks, such as visual captioning, visual question answering, and visual commonsense reasoning, is an important area in artificial intelligence and continuously attracts the research community's attention. Despite the improvements in overall performance, classic challenges still exist in vision-language tasks and hinder the development of this area. In recent years, the rise of pre-trained models is driving the research on vision-language tasks. Thanks to the massive scale of training data and model parameters, pre-trained models have exhibited excellent performance in numerous downstream tasks. Inspired by the powerful capabilities of pre-trained models, new paradigms have emerged to solve the classic challenges. Such methods have become mainstream in current research with increasing attention and rapid advances. In this paper, we present a comprehensive overview of how vision-language tasks benefit from pre-trained models. First, we review several main challenges in vision-language tasks and discuss the limitations of previous solutions before the era of pre-training. Next, we summarize the recent advances in incorporating pre-trained models to address the challenges in vision-language tasks. Finally, we analyze the potential risks associated with the inherent limitations of pre-trained models and discuss possible solutions, attempting to provide future research directions., Comment: Under Review

Vector set search, an underexplored similarity search paradigm, aims to find vector sets similar to a query set. This search paradigm leverages the inherent structural alignment between sets and real-world entities to model more fine-grained and consistent relationships for diverse applications. This task, however, faces more severe efficiency challenges than traditional single-vector search due to the combinatorial explosion of pairings in set-to-set comparisons. In this work, we aim to address the efficiency challenges posed by the combinatorial explosion in vector set search, as well as the curse of dimensionality inherited from single-vector search. To tackle these challenges, we present an efficient algorithm for vector set search, BioVSS (Bio-inspired Vector Set Search). BioVSS simulates the fly olfactory circuit to quantize vectors into sparse binary codes and then designs an index based on the set membership property of the Bloom filter. The quantization and indexing strategy enables BioVSS to efficiently perform vector set search by pruning the search space. Experimental results demonstrate over 50 times speedup compared to linear scanning on million-scale datasets while maintaining a high recall rate of up to 98.9%, making it an efficient solution for vector set search.

Nodes in the real-world graphs exhibit diverse patterns in numerous aspects, such as degree and homophily. However, most existent node predictors fail to capture a wide range of node patterns or to make predictions based on distinct node patterns, resulting in unsatisfactory classification performance. In this paper, we reveal that different node predictors are good at handling nodes with specific patterns and only apply one node predictor uniformly could lead to suboptimal result. To mitigate this gap, we propose a mixture of experts framework, MoE-NP, for node classification. Specifically, MoE-NP combines a mixture of node predictors and strategically selects models based on node patterns. Experimental results from a range of real-world datasets demonstrate significant performance improvements from MoE-NP.

Recent advances in imitation learning have shown significant promise for robotic control and embodied intelligence. However, achieving robust generalization across diverse mounted camera observations remains a critical challenge. In this paper, we introduce a video-based spatial perception framework that leverages 3D spatial representations to address environmental variability, with a focus on handling lighting changes. Our approach integrates a novel image augmentation technique, AugBlender, with a state-of-the-art monocular depth estimation model trained on internet-scale data. Together, these components form a cohesive system designed to enhance robustness and adaptability in dynamic scenarios. Our results demonstrate that our approach significantly boosts the success rate across diverse camera exposures, where previous models experience performance collapse. Our findings highlight the potential of video-based spatial perception models in advancing robustness for end-to-end robotic learning, paving the way for scalable, low-cost solutions in embodied intelligence., Comment: 8 pages, 5 figures

Drone swarms coupled with data intelligence can be the future of wildfire fighting. However, drone swarm firefighting faces enormous challenges, such as the highly complex environmental conditions in wildfire scenes, the highly dynamic nature of wildfire spread, and the significant computational complexity of drone swarm operations. We develop a predict-then-optimize approach to address these challenges to enable effective drone swarm firefighting. First, we construct wildfire spread prediction convex neural network (Convex-NN) models based on real wildfire data. Then, we propose a mixed-integer programming (MIP) model coupled with dynamic programming (DP) to enable efficient drone swarm task planning. We further use chance-constrained robust optimization (CCRO) to ensure robust firefighting performances under varying situations. The formulated model is solved efficiently using Benders Decomposition and Branch-and-Cut algorithms. After 75 simulated wildfire environments training, the MIP+CCRO approach shows the best performance among several testing sets, reducing movements by 37.3\% compared to the plain MIP. It also significantly outperformed the GA baseline, which often failed to fully extinguish the fire. Eventually, we will conduct real-world fire spread and quenching experiments in the next stage for further validation.

UAV remote sensing technology has become a key technology in crop breeding, which can achieve high-throughput and non-destructive collection of crop phenotyping data. However, the multidisciplinary nature of breeding has brought technical barriers and efficiency challenges to knowledge mining. Therefore, it is important to develop a smart breeding goal tool to mine cross-domain multimodal data. Based on different pre-trained open-source multimodal large language models (MLLMs) (e.g., Qwen-VL, InternVL, Deepseek-VL), this study used supervised fine-tuning (SFT), retrieval-augmented generation (RAG), and reinforcement learning from human feedback (RLHF) technologies to inject cross-domain knowledge into MLLMs, thereby constructing multiple multimodal large language models for wheat breeding (WBLMs). The above WBLMs were evaluated using the newly created evaluation benchmark in this study. The results showed that the WBLM constructed using SFT, RAG and RLHF technologies and InternVL2-8B has leading performance. Then, subsequent experiments were conducted using the WBLM. Ablation experiments indicated that the combination of SFT, RAG, and RLHF technologies can improve the overall generation performance, enhance the generated quality, balance the timeliness and adaptability of the generated answer, and reduce hallucinations and biases. The WBLM performed best in wheat yield prediction using cross-domain data (remote sensing, phenotyping, weather, germplasm) simultaneously, with R2 and RMSE of 0.821 and 489.254 kg/ha, respectively. Furthermore, the WBLM can generate professional decision support answers for phenotyping estimation, environmental stress assessment, target germplasm screening, cultivation technique recommendation, and seed price query tasks.

Rapid development of artificial intelligence has drastically accelerated the development of scientific discovery. Trained with large-scale observation data, deep neural networks extract the underlying patterns in an end-to-end manner and assist human researchers with highly-precised predictions in unseen scenarios. The recent rise of Large Language Models (LLMs) and the empowered autonomous agents enable scientists to gain help through interaction in different stages of their research, including but not limited to literature review, research ideation, idea implementation, and academic writing. However, AI researchers instantiated by foundation model empowered agents with full-process autonomy are still in their infancy. In this paper, we study $\textbf{AI-Generated Science}$ (AIGS), where agents independently and autonomously complete the entire research process and discover scientific laws. By revisiting the definition of scientific research, we argue that $\textit{falsification}$ is the essence of both human research process and the design of an AIGS system. Through the lens of falsification, prior systems attempting towards AI-Generated Science either lack the part in their design, or rely heavily on existing verification engines that narrow the use in specialized domains. In this work, we propose Baby-AIGS as a baby-step demonstration of a full-process AIGS system, which is a multi-agent system with agents in roles representing key research process. By introducing FalsificationAgent, which identify and then verify possible scientific discoveries, we empower the system with explicit falsification. Experiments on three tasks preliminarily show that Baby-AIGS could produce meaningful scientific discoveries, though not on par with experienced human researchers. Finally, we discuss on the limitations of current Baby-AIGS, actionable insights, and related ethical issues in detail., Comment: Pre-print. 35 pages. Official website: https://agent-force.github.io/AIGS/

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

In this paper, we introduce Hunyuan-Large, which is currently the largest open-source Transformer-based mixture of experts model, with a total of 389 billion parameters and 52 billion activation parameters, capable of handling up to 256K tokens. We conduct a thorough evaluation of Hunyuan-Large's superior performance across various benchmarks including language understanding and generation, logical reasoning, mathematical problem-solving, coding, long-context, and aggregated tasks, where it outperforms LLama3.1-70B and exhibits comparable performance when compared to the significantly larger LLama3.1-405B model. Key practice of Hunyuan-Large include large-scale synthetic data that is orders larger than in previous literature, a mixed expert routing strategy, a key-value cache compression technique, and an expert-specific learning rate strategy. Additionally, we also investigate the scaling laws and learning rate schedule of mixture of experts models, providing valuable insights and guidances for future model development and optimization. The code and checkpoints of Hunyuan-Large are released to facilitate future innovations and applications. Codes: https://github.com/Tencent/Hunyuan-Large Models: https://huggingface.co/tencent/Tencent-Hunyuan-Large, Comment: 17 pages, 4 Figures

Recommender systems play a pivotal role across practical scenarios, showcasing remarkable capabilities in user preference modeling. However, the centralized learning paradigm predominantly used raises serious privacy concerns. The federated recommender system (FedRS) addresses this by updating models on clients, while a central server orchestrates training without accessing private data. Existing FedRS approaches, however, face unresolved challenges, including non-convex optimization, vulnerability, potential privacy leakage risk, and communication inefficiency. This paper addresses these challenges by reformulating the federated recommendation problem as a convex optimization issue, ensuring convergence to the global optimum. Based on this, we devise a novel method, RFRec, to tackle this optimization problem efficiently. In addition, we propose RFRecF, a highly efficient version that incorporates non-uniform stochastic gradient descent to improve communication efficiency. In user preference modeling, both methods learn local and global models, collaboratively learning users' common and personalized interests under the federated learning setting. Moreover, both methods significantly enhance communication efficiency, robustness, and privacy protection, with theoretical support. Comprehensive evaluations on four benchmark datasets demonstrate RFRec and RFRecF's superior performance compared to diverse baselines., Comment: CIKM 2024

Transformer models have achieved remarkable success in sequential recommender systems (SRSs). However, computing the attention matrix in traditional dot-product attention mechanisms results in a quadratic complexity with sequence lengths, leading to high computational costs for long-term sequential recommendation. Motivated by the above observation, we propose a novel L2-Normalized Linear Attention for the Transformer-based Sequential Recommender Systems (LinRec), which theoretically improves efficiency while preserving the learning capabilities of the traditional dot-product attention. Specifically, by thoroughly examining the equivalence conditions of efficient attention mechanisms, we show that LinRec possesses linear complexity while preserving the property of attention mechanisms. In addition, we reveal its latent efficiency properties by interpreting the proposed LinRec mechanism through a statistical lens. Extensive experiments are conducted based on two public benchmark datasets, demonstrating that the combination of LinRec and Transformer models achieves comparable or even superior performance than state-of-the-art Transformer-based SRS models while significantly improving time and memory efficiency., Comment: SIGIR 2023

In this report, we propose a novel black-box jailbreak attacking framework that incorporates various LLM-as-Attacker methods to deliver transferable and powerful jailbreak attacks. Our method is designed based on three key observations from existing jailbreaking studies and practices. First, we consider an ensemble approach should be more effective in exposing the vulnerabilities of an aligned LLM compared to individual attacks. Second, different malicious instructions inherently vary in their jailbreaking difficulty, necessitating differentiated treatment to ensure more efficient attacks. Finally, the semantic coherence of a malicious instruction is crucial for triggering the defenses of an aligned LLM; therefore, it must be carefully disrupted to manipulate its embedding representation, thereby increasing the jailbreak success rate. We validated our approach by participating in the Competition for LLM and Agent Safety 2024, where our team achieved top performance in the Jailbreaking Attack Track.

We introduce motion graph, a novel approach to the video prediction problem, which predicts future video frames from limited past data. The motion graph transforms patches of video frames into interconnected graph nodes, to comprehensively describe the spatial-temporal relationships among them. This representation overcomes the limitations of existing motion representations such as image differences, optical flow, and motion matrix that either fall short in capturing complex motion patterns or suffer from excessive memory consumption. We further present a video prediction pipeline empowered by motion graph, exhibiting substantial performance improvements and cost reductions. Experiments on various datasets, including UCF Sports, KITTI and Cityscapes, highlight the strong representative ability of motion graph. Especially on UCF Sports, our method matches and outperforms the SOTA methods with a significant reduction in model size by 78% and a substantial decrease in GPU memory utilization by 47%., Comment: Accepted by NeurIPS 2024, 19 pages, 12 figures

Sim-to-real transfer remains a significant challenge in robotics due to the discrepancies between simulated and real-world dynamics. Traditional methods like Domain Randomization often fail to capture fine-grained dynamics, limiting their effectiveness for precise control tasks. In this work, we propose a novel approach that dynamically adjusts simulation environment parameters online using in-context learning. By leveraging past interaction histories as context, our method adapts the simulation environment dynamics to real-world dynamics without requiring gradient updates, resulting in faster and more accurate alignment between simulated and real-world performance. We validate our approach across two tasks: object scooping and table air hockey. In the sim-to-sim evaluations, our method significantly outperforms the baselines on environment parameter estimation by 80% and 42% in the object scooping and table air hockey setups, respectively. Furthermore, our method achieves at least 70% success rate in sim-to-real transfer on object scooping across three different objects. By incorporating historical interaction data, our approach delivers efficient and smooth system identification, advancing the deployment of robots in dynamic real-world scenarios. Demos are available on our project page: https://sim2real-capture.github.io/, Comment: website: https://sim2real-capture.github.io/

Massive scalar fields are promising candidates to address many unresolved problems in fundamental physics. We report the first model-agnostic Bayesian search of massive scalar fields in LIGO/Virgo/KAGRA gravitational-wave data. We find no evidence for such fields and place the most stringent upper limits on their coupling for scalar masses $\lesssim 2\times10^{-12}\,{\rm eV}$. We exemplify the strength of these bounds by applying them to massive scalar-Gauss-Bonnet gravity, finding the tightest constraints on the coupling constant to date, $\sqrt{\alpha_{\rm GB}}\lesssim 1\,{\rm km}$ for scalar masses $\lesssim 10^{-13}\,{\rm eV}$ to 90% confidence., Comment: 18 pages, 7 figures

The process of collecting blood from donors and making it available for transfusion requires a complex series of operations involving multiple actors and resources at each step. Ensuring hospitals receive adequate and safe blood for transfusion is a common challenge across low- and middle-income countries, but is rarely addressed from a system level. This paper presents the first use of discrete event simulation to study the blood system in Kenya and to explore the effect of variations and perturbations at different steps of the system on meeting patient blood demand. A process map of the Kenyan blood system was developed to capture critical steps from blood donation to transfusion using interviews with blood bank, hospital, and laboratory personnel at four public hospitals across three counties in Kenya. The blood system was simulated starting with blood collection, a blood bank where blood is tested and stored before it is issued, a major hospital attached to the blood bank, and several smaller hospitals served by the same blood bank. Values for supply-side parameters were based mainly on expert opinion; demand-side parameters were based on data from blood requisitions made in hospital wards, and dispatch of blood from the hospital laboratory. Illustrative examples demonstrate how the model can be used to explore the impacts of changes in blood collection (e.g., prioritising different donor types), blood demand (e.g., differing clinical case mix), and blood distribution (e.g., restocking strategies) on meeting demand at patient level. The model can reveal potential process impediments in the blood system and aid in choosing strategies for improving blood collection, distribution or use. Such a systems approach allows for interventions at different steps in the blood continuum to be tested on blood availability for different patients presenting at diverse hospitals across the country., Comment: 38 pages, 8 figures

The two-dimensional (2D) homogeneous electron gas (HEG) is a fundamental model in quantum many-body physics. It is important to theoretical and computational studies, where exchange-correlation energies computed in it serve as the foundation for density-functional calculations. It is also of direct relevance to a variety of experimental settings, especially with the rapid recent growth in 2D materials and moir\'e systems. In these experiments, metallic gates are often present, which screen the Coulomb interaction between electrons. The effect of the screening can qualitatively change the behavior of the 2D HEG, and requires accurate many-body computations to capture. In this work, we perform state-of-the-art diffusion Monte Carlo (DMC) calculations in the 2D HEG subjected to symmetric dual-gate screening. We systematically compute the correlation energy across a range of densities and gate separations for both spin unpolarized and fully polarized systems. A global fit is obtained for the correlation energy, using these data and imposing various limiting behaviors obtained from perturbation analysis. The new functional will allow density-functional calculations to be performed for a variety of realistic experimental setups which can accurately account for the presence of gates. We also investigate how the gate screening affects the bulk modulus, pair correlation function, and the structure factor of the 2D HEG, which can potentially be probed in experiments., Comment: 11 pages, 7 figures in the main text

Toroidal vortex is a kind of novel and exotic structured light with potential applications in photonic topology and quantum information. Here, we present the study on high harmonic generation (HHG) from the interaction of the intense toroidal vortices with atoms. We show that the spatial distribution of harmonic spectra reveal unique structures, which are kinds of high-order topological toroidal vortex because of the rotating of the driving fields with transverse orbital angular momentum. Then we show that, by synthesizing the toroidal vortices and optical vortices with longitudinal orbital angular momentum, it is able to generate the optical hopfions in extreme ultraviolet (EUV) range, which exhibit completely different topological property compared with both the driving fields. Harnessing the spin angular momentum selection rules, we further show that one can control the topological texture and channel each harmonic into a single mode of controllable Hopf invariant.

We discover a novel class of dissipative light bullets whose spatial profiles may be drastically reshaped along the bulk, edges and corners of the Su-Schrieffer-Heeger lattice owing to the dissipative and topological nature of the system. These light bullets appear due to resonances with different modes in lattice spectrum and may have very rich shapes that can be adiabatically controlled by varying the frequency of the external laser source. We report on robust stationary bullets and breathers which are understood from the corresponding bifurcation analysis. Our results provide a new route to realisation of reconfigurable and robust 3D light forms in topologically nontrivial systems., Comment: 7 pages, 4 figures, comments welcome

The recent wave of artificial intelligence, epitomized by large language models (LLMs), has presented opportunities and challenges for methodological innovation in political science, sparking discussions on a potential paradigm shift in the social sciences. However, how can we understand the impact of LLMs on knowledge production and paradigm transformation in the social sciences from a comprehensive perspective that integrates technology and methodology? What are LLMs' specific applications and representative innovative methods in political science research? These questions, particularly from a practical methodological standpoint, remain underexplored. This paper proposes the "Intelligent Computing Social Modeling" (ICSM) method to address these issues by clarifying the critical mechanisms of LLMs. ICSM leverages the strengths of LLMs in idea synthesis and action simulation, advancing intellectual exploration in political science through "simulated social construction" and "simulation validation." By simulating the U.S. presidential election, this study empirically demonstrates the operational pathways and methodological advantages of ICSM. By integrating traditional social science paradigms, ICSM not only enhances the quantitative paradigm's capability to apply big data to assess the impact of factors but also provides qualitative paradigms with evidence for social mechanism discovery at the individual level, offering a powerful tool that balances interpretability and predictability in social science research. The findings suggest that LLMs will drive methodological innovation in political science through integration and improvement rather than direct substitution., Comment: 34 pages, 5 figures, 3 tables

In this paper, we present a method, VectorPD, for converting a given human face image into a vector portrait sketch. VectorPD supports different levels of abstraction by simply controlling the number of strokes. Since vector graphics are composed of different shape primitives, it is challenging for rendering complex faces to accurately express facial details and structure. To address this, VectorPD employs a novel two-round optimization mechanism. We first initialize the strokes with facial keypoints, and generate a basic portrait sketch by a CLIP-based Semantic Loss. Then we complete the face structure through VGG-based Structure Loss, and propose a novel Crop-based Shadow Loss to enrich the shadow details of the sketch, achieving a visually pleasing portrait sketch. Quantitative and qualitative evaluations both demonstrate that the portrait sketches generated by VectorPD can produce better visual effects than existing state-of-the-art methods, maintaining as much fidelity as possible at different levels of abstraction., Comment: 9 pages, 12 figures

Scene sketching is to convert a scene into a simplified, abstract representation that captures the essential elements and composition of the original scene. It requires semantic understanding of the scene and consideration of different regions within the scene. Since scenes often contain diverse visual information across various regions, such as foreground objects, background elements, and spatial divisions, dealing with these different regions poses unique difficulties. In this paper, we define a sketch as some sets of Bezier curves. We optimize the different regions of input scene in multiple rounds. In each round of optimization, strokes sampled from the next region can seamlessly be integrated into the sketch generated in the previous round of optimization. We propose additional stroke initialization method to ensure the integrity of the scene and the convergence of optimization. A novel CLIP-Based Semantic loss and a VGG-Based Feature loss are utilized to guide our multi-round optimization. Extensive experimental results on the quality and quantity of the generated sketches confirm the effectiveness of our method., Comment: 9 pages, 9 figures

Pre-training on large-scale, high-quality datasets is crucial for enhancing the reasoning capabilities of Large Language Models (LLMs), especially in specialized domains such as mathematics. Despite the recognized importance, the Multimodal LLMs (MLLMs) field currently lacks a comprehensive open-source pre-training dataset specifically designed for mathematical reasoning. To address this gap, we introduce InfiMM-WebMath-40B, a high-quality dataset of interleaved image-text documents. It comprises 24 million web pages, 85 million associated image URLs, and 40 billion text tokens, all meticulously extracted and filtered from CommonCrawl. We provide a detailed overview of our data collection and processing pipeline. To demonstrate the robustness of InfiMM-WebMath-40B, we conducted evaluations in both text-only and multimodal settings. Our evaluations on text-only benchmarks show that, despite utilizing only 40 billion tokens, our dataset significantly enhances the performance of our 1.3B model, delivering results comparable to DeepSeekMath-1.3B, which uses 120 billion tokens for the same model size. Nevertheless, with the introduction of our multi-modal math pre-training dataset, our models set a new state-of-the-art among open-source models on multi-modal math benchmarks such as MathVerse and We-Math. We release our data at https://huggingface.co/datasets/Infi-MM/InfiMM-WebMath-40B.

Transformer uses GPU as the initial design platform, but GPU can only perform limited hardware customization. Although FPGA has strong customization ability, the design solution space is huge and the design difficulty is high. Versal ACAP is a heterogeneous computing architecture with AI Engine as the core. It is far more flexible than GPU in hardware customization, and has better and smaller design solution space than traditional FPGA. Therefore, this paper proposes the Customized Transformer Accelerator Framework(CAT), through the CAT framework, a customized Transformer accelerator family can be derived on Versal ACAP, CAT framework has an abstract accelerator architecture design idea, which deconstructs and efficiently maps the Transformer into the hardware, which contains a variety of customizable properties. Through the customization and optimization strategy of the CAT framework, the underlying hardware and the upper model jointly constrain and decide on these customizable properties, and finally form a customized accelerator. We use a 7 nm AMD Versal ACAP VCK5000 development board to implement accelerators for different Transformer models based on the CAT framework. Experiments show that we achieve the highest throughput gains of 2.41x, 49.50x, and 1.32x compared to 8 nm Nvidia GPU A10G, 16 nm AMD FPGA ZCU102, and 7 nm AMD Versal ACAP VC190(SOTA). The highest energy efficiency gains are 7.80x, 6.19x and 1.15x, respectively.

Indoor robots rely on depth to perform tasks like navigation or obstacle detection, and single-image depth estimation is widely used to assist perception. Most indoor single-image depth prediction focuses less on model generalizability to unseen datasets, concerned with in-the-wild robustness for system deployment. This work leverages gradient-based meta-learning to gain higher generalizability on zero-shot cross-dataset inference. Unlike the most-studied meta-learning of image classification associated with explicit class labels, no explicit task boundaries exist for continuous depth values tied to highly varying indoor environments regarding object arrangement and scene composition. We propose fine-grained task that treats each RGB-D mini-batch as a task in our meta-learning formulation. We first show that our method on limited data induces a much better prior (max 27.8% in RMSE). Then, finetuning on meta-learned initialization consistently outperforms baselines without the meta approach. Aiming at generalization, we propose zero-shot cross-dataset protocols and validate higher generalizability induced by our meta-initialization, as a simple and useful plugin to many existing depth estimation methods. The work at the intersection of depth and meta-learning potentially drives both research to step closer to practical robotic and machine perception usage., Comment: IROS 2024. The version supersedes 2305.07269. arXiv admin note: text overlap with arXiv:2305.07269

In this survey, we design formal verification and control algorithms for autonomous systems with practical safety guarantees using conformal prediction (CP), a statistical tool for uncertainty quantification. We focus on learning-enabled autonomous systems (LEASs) in which the complexity of learning-enabled components (LECs) is a major bottleneck that hampers the use of existing model-based verification and design techniques. Instead, we advocate for the use of CP, and we will demonstrate its use in formal verification, systems and control theory, and robotics. We argue that CP is specifically useful due to its simplicity (easy to understand, use, and modify), generality (requires no assumptions on learned models and data distributions, i.e., is distribution-free), and efficiency (real-time capable and accurate). We pursue the following goals with this survey. First, we provide an accessible introduction to CP for non-experts who are interested in using CP to solve problems in autonomy. Second, we show how to use CP for the verification of LECs, e.g., for verifying input-output properties of neural networks. Third and fourth, we review recent articles that use CP for safe control design as well as offline and online verification of LEASs. We summarize their ideas in a unifying framework that can deal with the complexity of LEASs in a computationally efficient manner. In our exposition, we consider simple system specifications, e.g., robot navigation tasks, as well as complex specifications formulated in temporal logic formalisms. Throughout our survey, we compare to other statistical techniques (e.g., scenario optimization, PAC-Bayes theory, etc.) and how these techniques have been used in verification and control. Lastly, we point the reader to open problems and future research directions.

Through a review of current game practices, the author highlights concerns regarding the safety of public social games and the singular medical approach to serious game design for autism. The paper identifies a disconnect between the needs of autistic children and the existing solutions. To fill this gap, a neurodiversity approach to serious game design is proposed. This approach aims to address the social needs of autistic children, enabling them to interact with their neurotypical peers directly, confidently, and safely., Comment: This paper has been accepted by the Joint Conference on Serious Games 2024 and will be published in November 2024 by Springer as part of their LNCS series

This ongoing experimental project investigates the use of Generative Image Models (GIMs) in crafting a picture book creation game designed to nurture social connections among autistic children and their neurotypical peers within a neuro-affirming environment. Moving away from traditional methods that often seek to condition neurodivergent children to socialize in prescribed ways, this project strives to cultivate a space where children can engage with one another naturally and creatively through art and storytelling, free from the pressure to adhere to standard social norms. Beyond merely "story-choosing," the research highlights the potential of GIMs to facilitate "story-creating," fostering peer social connections in a creative and structured collaborative learning experience., Comment: This paper has been accepted by the Joint Conference on Serious Games 2024 and will be published in November 2024 by Springer as part of their LNCS series

We report a lattice QCD calculation of the parton distribution function (PDF) of a deuteron-like dibaryon system using large-momentum effective theory. The calculation is done on three Wilson Clover ensembles with a fixed lattice spacing a=0.105 fm and two pion masses. The lattice matrix elements are computed at proton momenta up to 2.46 GeV with the signal of high momentum modes being improved by applying the momentum smearing technique. The state-of-the-art renormalization, matching and extrapolation are then applied to obtain the final result of the light-cone PDF. A comparison between the result of the dibaryon system and the sum of the proton and neutron PDFs is also given.

$\pi$ modes are unique topological edge states appearing in Floquet systems with periodic modulations of the underlying lattice structure in evolution variable, such as dynamically modulated Su-Schrieffer-Heeger (SSH) lattices. These edge states are anomalous states usually appearing between Floquet replicas of the same band, even if standard topological index remains zero for this band. While linear and nonlinear $\pi$ modes were observed in conservative systems, they have never been studied in nonlinear regime in the non-Hermitian systems with structured gain and losses. Here we show that SSH waveguide array with periodically oscillating waveguide positions in propagation direction and with parity-time symmetric refractive index landscape, can support $\pi$ modes that are damped or amplified at different ends of the array. By including nonlinearity and nonlinear absorption into our continuous system, we achieve stable lasing in $\pi$ mode at one end of the array. The representative feature of this system is that lasing in it is thresholdless and it occurs even at low gain-loss amplitudes. The degree of localization of lasing $\pi$ modes can be flexibly controlled by the amplitude of transverse waveguide oscillations. This work therefore introduces a new type of topological Floquet laser and a route to manipulation of $\pi$ modes by structured gain and losses., Comment: 9 pages, 3 figures, to appear in APL Photonics

Top-k Nearest Neighbors (kNN) problem on road network has numerous applications on location-based services. As direct search using the Dijkstra's algorithm results in a large search space, a plethora of complex-index-based approaches have been proposed to speedup the query processing. However, even with the current state-of-the-art approach, long query processing delays persist, along with significant space overhead and prohibitively long indexing time. In this paper, we depart from the complex index designs prevalent in existing literature and propose a simple index named KNN-Index. With KNN-Index, we can answer a kNN query optimally and progressively with small and size-bounded index. To improve the index construction performance, we propose a bidirectional construction algorithm which can effectively share the common computation during the construction. Theoretical analysis and experimental results on real road networks demonstrate the superiority of KNN-Index over the state-of-the-art approach in query processing performance, index size, and index construction efficiency., Comment: 15 pages, 15 figures

As AI chips incorporate numerous parallelized cores to scale deep learning (DL) computing, inter-core communication is enabled recently by employing high-bandwidth and low-latency interconnect links on the chip (e.g., Graphcore IPU). It allows each core to directly access the fast scratchpad memory in other cores, which enables new parallel computing paradigms. However, without proper support for the scalable inter-core connections in current DL compilers, it is hard for developers to exploit the benefits of this new architecture. We present T10, the first DL compiler to exploit the inter-core communication bandwidth and distributed on-chip memory on AI chips. To formulate the computation and communication patterns of tensor operators in this new architecture, T10 introduces a distributed tensor abstraction rTensor. T10 maps a DNN model to execution plans with a generalized compute-shift pattern, by partitioning DNN computation into sub-operators and mapping them to cores, so that the cores can exchange data following predictable patterns. T10 makes globally optimized trade-offs between on-chip memory consumption and inter-core communication overhead, selects the best execution plan from a vast optimization space, and alleviates unnecessary inter-core communications. Our evaluation with a real inter-core connected AI chip, the Graphcore IPU, shows up to 3.3$\times$ performance improvement, and scalability support for larger models, compared to state-of-the-art DL compilers and vendor libraries., Comment: This paper is accepted at The 30th ACM Symposium on Operating Systems Principles (SOSP'24)

Cloud platforms today have been deploying hardware accelerators like neural processing units (NPUs) for powering machine learning (ML) inference services. To maximize the resource utilization while ensuring reasonable quality of service, a natural approach is to virtualize NPUs for efficient resource sharing for multi-tenant ML services. However, virtualizing NPUs for modern cloud platforms is not easy. This is not only due to the lack of system abstraction support for NPU hardware, but also due to the lack of architectural and ISA support for enabling fine-grained dynamic operator scheduling for virtualized NPUs. We present Neu10, a holistic NPU virtualization framework. We investigate virtualization techniques for NPUs across the entire software and hardware stack. Neu10 consists of (1) a flexible NPU abstraction called vNPU, which enables fine-grained virtualization of the heterogeneous compute units in a physical NPU (pNPU); (2) a vNPU resource allocator that enables pay-as-you-go computing model and flexible vNPU-to-pNPU mappings for improved resource utilization and cost-effectiveness; (3) an ISA extension of modern NPU architecture for facilitating fine-grained tensor operator scheduling for multiple vNPUs. We implement Neu10 based on a production-level NPU simulator. Our experiments show that Neu10 improves the throughput of ML inference services by up to 1.4$\times$ and reduces the tail latency by up to 4.6$\times$, while improving the NPU utilization by 1.2$\times$ on average, compared to state-of-the-art NPU sharing approaches., Comment: Accepted to MICRO'24

Deep learning has been widely applied in recommender systems, which has achieved revolutionary progress recently. However, most existing learning-based methods assume that the user and item distributions remain unchanged between the training phase and the test phase. However, the distribution of user and item features can naturally shift in real-world scenarios, potentially resulting in a substantial decrease in recommendation performance. This phenomenon can be formulated as an Out-Of-Distribution (OOD) recommendation problem. To address this challenge, we propose a novel Dual Test-Time-Training framework for OOD Recommendation, termed DT3OR. In DT3OR, we incorporate a model adaptation mechanism during the test-time phase to carefully update the recommendation model, allowing the model to specially adapt to the shifting user and item features. To be specific, we propose a self-distillation task and a contrastive task to assist the model learning both the user's invariant interest preferences and the variant user/item characteristics during the test-time phase, thus facilitating a smooth adaptation to the shifting features. Furthermore, we provide theoretical analysis to support the rationale behind our dual test-time training framework. To the best of our knowledge, this paper is the first work to address OOD recommendation via a test-time-training strategy. We conduct experiments on three datasets with various backbones. Comprehensive experimental results have demonstrated the effectiveness of DT3OR compared to other state-of-the-art baselines.

Flat-band periodic materials are characterized by a linear spectrum containing at least one band where the propagation constant remains nearly constant irrespective of the Bloch momentum across the Brillouin zone. These materials provide a unique platform for investigating phenomena related to light localization. Meantime, the interaction between flat-band physics and nonlinearity in continuous systems remains largely unexplored, particularly in continuous systems where the band flatness deviates slightly from zero, in contrast to simplified discrete systems with exactly flat bands. Here, we use a continuous superhoneycomb lattice featuring a flat band in its spectrum to theoretically and numerically introduce a range of stable flatband solitons. These solutions encompass fundamental, dipole, multi-peak, and even vortex solitons. Numerical analysis demonstrates that these solitons are stable in a broad range of powers. They do not bifurcate from the flat band and can be analyzed using Wannier function expansion leading to their designation as Wannier solitons. These solitons showcase novel possibilities for light localization and transmission within nonlinear flat-band systems., Comment: 12 pages, 5 figures, to appear in Nanophotonics. Comments are welcome

The goal of stock trend prediction is to forecast future market movements for informed investment decisions. Existing methods mostly focus on predicting stock trends with supervised models trained on extensive annotated data. However, human annotation can be resource-intensive and the annotated data are not readily available. Inspired by the impressive few-shot capability of Large Language Models (LLMs), we propose using LLMs in a few-shot setting to overcome the scarcity of labeled data and make prediction more feasible to investors. Previous works typically merge multiple financial news for predicting stock trends, causing two significant problems when using LLMs: (1) Merged news contains noise, and (2) it may exceed LLMs' input limits, leading to performance degradation. To overcome these issues, we propose a two-step method 'denoising-then-voting'. Specifically, we introduce an `Irrelevant' category, and predict stock trends for individual news instead of merged news. Then we aggregate these predictions using majority voting. The proposed method offers two advantages: (1) Classifying noisy news as irrelevant removes its impact on the final prediction. (2) Predicting for individual news mitigates LLMs' input length limits. Our method achieves 66.59% accuracy in S&P 500, 62.17% in CSI-100, and 61.17% in HK stock prediction, outperforming the standard few-shot counterparts by around 7%, 4%, and 4%. Furthermore, our proposed method performs on par with state-of-the-art supervised methods.

We propose a chiral bulk soliton in a nonlinear photonic lattice with decorated boundaries, presenting a novel approach to manipulate photonic transport without extensive bulk modifications. Unlike traditional methods that rely on topological edge and corner modes, our strategy leverages the robust chiral propagation of bulk modes. By introducing nonlinearity into the system, we find a stable bulk soliton, akin to the topological valley Hall effects. The chiral bulk soliton exhibits remarkable stability; the energy does not decay even after a long-distance propagation; and the corresponding Fourier spectrum confirms the absence of inter-valley scattering indicating a valley-locking property. Our findings not only contribute to the fundamental understanding of nonlinear photonic systems but also hold significant practical implications for the design and optimization of photonic devices.

The type-II Dirac cone is a special feature of the band structure, whose Fermi level is represented by a pair of crossing lines. It has been demonstrated that such a structure is useful for investigating topological edge solitons, and more specifically, for mimicking the Kline tunneling. However, it is still not clear what the interplay between type-II Dirac cones and the non-Hermiticity mechanism will result in. Here, this question is addressed; in particular, we report the $\mathcal{PT}$-symmetric photonic lattices with type-II Dirac cones for the first time. We identify a slope-exceptional ring and name it the type-II exceptional ring. We display the restoration of the $\mathcal{PT}$ symmetry of the lattice by reducing the separation between the sites in the unit cell. Curiously, the amplitude of the beam during propagation in the non-Hermitian lattice with $\mathcal{PT}$ symmetry only decays because of diffraction, whereas in the $\mathcal{PT}$ symmetry-broken lattice it will be amplified, even though the beam still diffracts. This work establishes the link between the non-Hermiticity mechanism and the violation of Lorentz invariance in these physical systems., Comment: 5 pages, 4 figures, to appear in Optics Letters. Comments are welcome

Topological insulators are most frequently constructed using lattices with specific degeneracies in their linear spectra, such as Dirac points. For a broad class of lattices, such as honeycomb ones, these points and associated Dirac cones generally appear in non-equivalent pairs. Simultaneous breakup of the time-reversal and inversion symmetry in systems based on such lattices may result in the formation of the unpaired Dirac cones in bulk spectrum, but the existence of topologically protected edge states in such structures remains an open problem. Here photonic Floquet insulator on honeycomb lattice with unpaired Dirac cones in its spectrum is introduced that can support unidirectional edge states appearing at the edge between two regions with opposite sublattice detuning. Topological properties of this system are characterized by the nonzero valley Chern number. Remarkably, edge states in this system can circumvent sharp corners without inter-valley scattering even though there is no total forbidden gap in the spectrum. Our results reveal unusual interplay between two different physical mechanisms of creation of topological edge states based on simultaneous breakup of different symmetries of the system., Comment: 9 pages, 7 figures. To appear in Photonics Research. Comments are welcome

Adiabatic mode evolution waveguides (AMEWs) are widely utilized in integrated photonics, including tapered waveguides, edge couplers, mode converters, splitters, etc. An analytical theory and a novel AMEW design algorithm are developed to create shortcuts to adiabaticity (STA). This new algorithm is effective in shortening the total length of the AMEW while maintaining the desired wavelength range. Moreover, this analytical algorithm requires much fewer computing resources than traditional numerical algorithms. With the new algorithm, we demonstrate a broadband and highly efficient polarization splitter-rotator (PSR) on a lithium-niobate-on-insulator (LNOI) platform with an LN thickness of 500 nm. According to our simulation, the length of the PSR is shortened by 3.5 times compared to the linear design. The fabricated PSR, with a total length of 2 mm, exhibits an insertion loss (IL) of 0.8 dB and a polarization extinction ratio (ER) of 12.2 dB over a wavelength range exceeding 76 nm., Comment: 16 pages, 6 figures, 2 tables

The emergence of the large language model~(LLM) poses an exponential growth of demand for computation throughput, memory capacity, and communication bandwidth. Such a demand growth has significantly surpassed the improvement of corresponding chip designs. With the advancement of fabrication and integration technologies, designers have been developing Wafer-Scale Chips~(WSCs) to scale up and exploit the limits of computation density, memory capacity, and communication bandwidth at the level of a single chip. Existing solutions have demonstrated the significant advantages of WSCs over traditional designs, showing potential to effectively support LLM workloads. Despite the benefits, exploring the early-stage design space of WSCs for LLMs is a crucial yet challenging task due to the enormous and complicated design space, time-consuming evaluation methods, and inefficient exploration strategies. To address these challenges, we propose Theseus, an efficient WSC design space exploration framework for LLMs. We construct the design space of WSCs with various constraints considering the unique characteristics of WSCs. We propose efficient evaluation methodologies for large-scale NoC-based WSCs and introduce multi-fidelity Bayesian optimization to efficiently explore the design space. Evaluation results demonstrate the efficiency of Theseus that the searched Pareto optimal results outperform GPU cluster and existing WSC designs by up to 62.8\%/73.7\% in performance and 38.6\%/42.4\% in power consumption for LLM training, while improving up to 23.2$\times$ and 15.7$\times$ for the performance and power of inference tasks. Furthermore, we conduct case studies to address the design tradeoffs in WSCs and provide insights to facilitate WSC designs for LLMs.

Animal ethology is an crucial aspect of animal research, and animal behavior labeling is the foundation for studying animal behavior. This process typically involves labeling video clips with behavioral semantic tags, a task that is complex, subjective, and multimodal. With the rapid development of multimodal large language models(LLMs), new application have emerged for animal behavior understanding tasks in livestock scenarios. This study evaluates the visual perception capabilities of multimodal LLMs in animal activity recognition. To achieve this, we created piglet test data comprising close-up video clips of individual piglets and annotated full-shot video clips. These data were used to assess the performance of four multimodal LLMs-Video-LLaMA, MiniGPT4-Video, Video-Chat2, and GPT-4 omni (GPT-4o)-in piglet activity understanding. Through comprehensive evaluation across five dimensions, including counting, actor referring, semantic correspondence, time perception, and robustness, we found that while current multimodal LLMs require improvement in semantic correspondence and time perception, they have initially demonstrated visual perception capabilities for animal activity recognition. Notably, GPT-4o showed outstanding performance, with Video-Chat2 and GPT-4o exhibiting significantly better semantic correspondence and time perception in close-up video clips compared to full-shot clips. The initial evaluation experiments in this study validate the potential of multimodal large language models in livestock scene video understanding and provide new directions and references for future research on animal behavior video understanding. Furthermore, by deeply exploring the influence of visual prompts on multimodal large language models, we expect to enhance the accuracy and efficiency of animal behavior recognition in livestock scenarios through human visual processing methods.

In this paper, we study the parabolic equations $\partial_t u=\partial_j\left(a^{ij}(x,t)\partial_iu\right)+b^j(x,t)\partial_ju+c(x,t)u$ in a domain of $\mathbb{R}^n$ under the condition that $a^{ij}$ are Lipschitz continuous. Consider the nodal set $Z_t=\{x: u(x,t)=0\}$ at a time $t$-slice. Simple examples show that the singular set $\mathcal{S}_t=\{x: u(x,t)=|\nabla_x u|(x,t)=0\}$ may coincide with nodal set. This makes the methods used in the study of nodal sets for elliptic equations fail, rendering the parabolic case much more complicated. The current strongest results in the literature establish the finiteness of the $(n-1)$-dimensional Hausdorff measure of $Z_t$, assuming either $n=1$ by Angenent or that the coefficients are time-independent and analytic by Lin. With general coefficients, the codimension-one estimate was obtained under some doubling assumption by Han-Lin but only for space-time nodal sets. In the first part, we prove that $\mathcal{H}^{n-1}(Z_t) < \infty$ in full generality, i.e. for any dimension, with time-dependent coefficients and with merely Lipschitz regular leading coefficients $a^{ij}$. In the second part, we study the evolutionary behavior of nodal sets. When $n=1$, it is proved by Angenent that the number of nodal points is non-increasing in time. For the $n$-dimensional case, we construct examples showing that measure monotonicity fails. In contrast, we prove dimension monotonicity, i.e., the Hausdorff dimension of the nodal set is non-increasing in time. This is the first monotonicity property for nodal sets in general dimensions. All the assumptions here are sharp.

May-Thurner Syndrome (MTS), also known as iliac vein compression syndrome or Cockett's syndrome, is a condition potentially impacting over 20 percent of the population, leading to an increased risk of iliofemoral deep venous thrombosis. In this paper, we present a 3D-based deep learning approach called MTS-Net for diagnosing May-Thurner Syndrome using CT scans. To effectively capture the spatial-temporal relationship among CT scans and emulate the clinical process of diagnosing MTS, we propose a novel attention module called the dual-enhanced positional multi-head self-attention (DEP-MHSA). The proposed DEP-MHSA reconsiders the role of positional embedding and incorporates a dual-enhanced positional embedding in both attention weights and residual connections. Further, we establish a new dataset, termed MTS-CT, consisting of 747 subjects. Experimental results demonstrate that our proposed approach achieves state-of-the-art MTS diagnosis results, and our self-attention design facilitates the spatial-temporal modeling. We believe that our DEP-MHSA is more suitable to handle CT image sequence modeling and the proposed dataset enables future research on MTS diagnosis. We make our code and dataset publicly available at: https://github.com/Nutingnon/MTS_dep_mhsa.

Training models with longer in-context lengths is a significant challenge for multimodal model due to substantial GPU memory and computational costs. This exploratory study does not present state-of-the-art models; rather, it introduces an innovative method designed to increase in-context text length in multi-modality large language models (MLLMs) efficiently. We present Visualized In-Context Text Processing (VisInContext), which processes long in-context text using visual tokens. This technique significantly reduces GPU memory usage and floating point operations (FLOPs) for both training and inferenceing stage. For instance, our method expands the pre-training in-context text length from 256 to 2048 tokens with nearly same FLOPs for a 56 billion parameter MOE model. Experimental results demonstrate that model trained with VisInContext delivers superior performance on common downstream benchmarks for in-context few-shot evaluation. Additionally, VisInContext is complementary to existing methods for increasing in-context text length and enhances document understanding capabilities, showing great potential in document QA tasks and sequential document retrieval., Comment: 12 pages. The website is \url{https://fingerrec.github.io/visincontext}

Knowledge of the domain of applicability of a machine learning model is essential to ensuring accurate and reliable model predictions. In this work, we develop a new approach of assessing model domain and demonstrate that our approach provides accurate and meaningful designation of in-domain versus out-of-domain when applied across multiple model types and material property data sets. Our approach assesses the distance between a test and training data point in feature space by using kernel density estimation and shows that this distance provides an effective tool for domain determination. We show that chemical groups considered unrelated based on established chemical knowledge exhibit significant dissimilarities by our measure. We also show that high measures of dissimilarity are associated with poor model performance (i.e., high residual magnitudes) and poor estimates of model uncertainty (i.e., unreliable uncertainty estimation). Automated tools are provided to enable researchers to establish acceptable dissimilarity thresholds to identify whether new predictions of their own machine learning models are in-domain versus out-of-domain., Comment: 43 pages, 34 figures, journal submission

Distribution-preserving integrated sensing and communication with secure reconstruction is investigated in this paper. In addition to the distortion constraint, we impose another constraint on the distance between the reconstructed sequence distribution and the original state distribution to force the system to preserve the statistical property of the channel states. An inner bound of the distribution-preserving capacity-distortion region is provided with some capacity region results under special cases. A numerical example demonstrates the tradeoff between the communication rate, reconstruction distortion and distribution preservation. Furthermore, we consider the case that the reconstructed sequence should be kept secret from an eavesdropper who also observes the channel output. An inner bound of the tradeoff region and a capacity-achieving special case are presented., Comment: Accepted by ISIT2024

We propose a high-performance glass-plastic hybrid minimalist aspheric panoramic annular lens (ASPAL) to solve several major limitations of the traditional panoramic annular lens (PAL), such as large size, high weight, and complex system. The field of view (FoV) of the ASPAL is 360{\deg}x(35{\deg}~110{\deg}) and the imaging quality is close to the diffraction limit. This large FoV ASPAL is composed of only 4 lenses. Moreover, we establish a physical structure model of PAL using the ray tracing method and study the influence of its physical parameters on compactness ratio. In addition, for the evaluation of local tolerances of annular surfaces, we propose a tolerance analysis method suitable for ASPAL. This analytical method can effectively analyze surface irregularities on annular surfaces and provide clear guidance on manufacturing tolerances for ASPAL. Benefiting from high-precision glass molding and injection molding aspheric lens manufacturing techniques, we finally manufactured 20 ASPALs in small batches. The weight of an ASPAL prototype is only 8.5 g. Our framework provides promising insights for the application of panoramic systems in space and weight-constrained environmental sensing scenarios such as intelligent security, micro-UAVs, and micro-robots., Comment: Accepted to Optics & Laser Technology