Search

Asian Americans as a group are overrepresented among STEM college graduates and have the highest average college enrollment rate of any racial or ethnic category. Thus, Asian Americans are typically excluded from educational interventions directed at improving STEM education for Students of Color because they are not considered to be underrepresented minorities. However, statistics obscure the individual needs of the more than 20 ethnic subgroups that fall under the umbrella term Asian Americans. Using a participatory action research approach, this paper documents the institutional and sociocultural factors that push out HMoob (or Hmong) American college students from STEM programs at one large, predominantly White university; and the coordinate processes of gatekeeping and transactional advising that either redirect those students toward non-STEM programs or force them out of the university completely.

Due to the success of the Standard Model~(SM), it is reasonable to anticipate that, the signal of new physics~(NP) beyond the SM is small, and future searches for NP and precision tests of the SM will require high luminosity collider experiments. Moreover, as the precision tests of the SM advances, rarer processes with a greater number of final-state particles will require consideration, which will in turn require the analysis of a multitude of observables. As an inherent consequence of the high luminosity, the generation of a large amount of experimental data in a large feature space presents a significant challenge for data processing. In recent years, quantum machine learning has emerged as a promising approach for processing large amounts of complex data on a quantum computer. In this study, we propose a variational quantum searching neighbor~(VQSN) algorithm to search for NP. As an example, we apply the VQSN in the phenomenological study of the gluon quartic gauge couplings~(gQGCs) at the Large Hadron Collider. The results suggest that VQSN demonstrates superior efficiency to a classical counterpart k-nearest neighbor algorithm, even when dealing with classical data., Comment: 18 pages, 1 figure

Recent advances in prototypical learning have shown remarkable potential to provide useful decision interpretations associating activation maps and predictions with class-specific training prototypes. Such prototypical learning has been well-studied for various single-label diseases, but for quite relevant and more challenging multi-label diagnosis, where multiple diseases are often concurrent within an image, existing prototypical learning models struggle to obtain meaningful activation maps and effective class prototypes due to the entanglement of the multiple diseases. In this paper, we present a novel Cross- and Intra-image Prototypical Learning (CIPL) framework, for accurate multi-label disease diagnosis and interpretation from medical images. CIPL takes advantage of common cross-image semantics to disentangle the multiple diseases when learning the prototypes, allowing a comprehensive understanding of complicated pathological lesions. Furthermore, we propose a new two-level alignment-based regularisation strategy that effectively leverages consistent intra-image information to enhance interpretation robustness and predictive performance. Extensive experiments show that our CIPL attains the state-of-the-art (SOTA) classification accuracy in two public multi-label benchmarks of disease diagnosis: thoracic radiography and fundus images. Quantitative interpretability results show that CIPL also has superiority in weakly-supervised thoracic disease localisation over other leading saliency- and prototype-based explanation methods.

CLIP is one of the most important multimodal foundational models today. What powers CLIP's capabilities? The rich supervision signals provided by natural language, the carrier of human knowledge, shape a powerful cross-modal representation space. However, with the rapid advancements in large language models LLMs like GPT-4 and LLaMA, the boundaries of language comprehension and generation are continually being pushed. This raises an intriguing question: can the capabilities of LLMs be harnessed to further improve multimodal representation learning? The potential benefits of incorporating LLMs into CLIP are clear. LLMs' strong textual understanding can fundamentally improve CLIP's ability to handle image captions, drastically enhancing its ability to process long and complex texts, a well-known limitation of vanilla CLIP. Moreover, LLMs are trained on a vast corpus of text, possessing open-world knowledge. This allows them to expand on caption information during training, increasing the efficiency of the learning process. In this paper, we propose LLM2CLIP, a novel approach that embraces the power of LLMs to unlock CLIP's potential. By fine-tuning the LLM in the caption space with contrastive learning, we extract its textual capabilities into the output embeddings, significantly improving the output layer's textual discriminability. We then design an efficient training process where the fine-tuned LLM acts as a powerful teacher for CLIP's visual encoder. Thanks to the LLM's presence, we can now incorporate longer and more complex captions without being restricted by vanilla CLIP's text encoder's context window and ability limitations. Our experiments demonstrate that this approach brings substantial improvements in cross-modal tasks.

This study investigates Neutral Triple Gauge Couplings (nTGCs) through $ZZ$ production at future electron-positron colliders. The impact of beam polarization on cross section is analyzed. We compare the signals and backgrounds for five different $ZZ$ decay channels and present our event selection strategies for future $e^+e^-$ colliders. The expected coefficient constraints for each decay channels are provided, and final expected constraints are derived by combining results from the different decay patterns. Our analysis indicates that future electron-positron colliders will have significantly enhanced detection capabilities for nTGCs compared to the current LHC experiments, with expected improvements in constraints by one to two orders of magnitude., Comment: 36 pages, 25 figures, 19 tables

Differential absorption Lidar (DIAL) in the ultraviolet (UV) region is an effective approach for monitoring tropospheric ozone. 4H-SiC single-photon detectors (SPDs) are emergent devices for UV single-photon detection. Here, we demonstrate a 4H-SiC SPD-based ozone DIAL. We design and fabricate the 4H-SiC single-photon avalanche diode with a beveled mesa structure and optimized layer thickness. An active quenching circuit with a quenching time of 1.03 ns is developed to significantly mitigate the afterpulsing effect while enhancing the maximum count rate. After characterization, the SPD exhibits excellent performance with a photon detection efficiency of 16.6% at 266 nm, a dark count rate of 138 kcps, a maximum count rate of 13 Mcps, and an afterpulse probability of 2.7% at room temperature. Then, we apply two 4H-SiC SPDs in an ozone DIAL. The measured ozone concentrations at altitudes of 1-3.5 km agree well with the results of a commercial ozone DIAL. Our work provides an alternative solution for general UV Lidar applications., Comment: Accepted by Applied Physics Letters

Cooperatively Localizing robots should seek optimal control strategies to maximize precision of position estimation and ensure safety in flight. Observability-Aware Trajectory Optimization has strong potential to address this issue, but no concrete link between observability and precision has been proven yet. In this paper, we prove that improvement in positioning precision inherently follows from optimizing observability. Based on this finding, we develop an Observability-Aware Control principle to generate observability-optimal control strategies. We implement this principle in a Model Predictive Control framework, and we verify it on a team of quadrotor Unmanned Aerial Vehicles comprising a follower vehicle localizing itself by tracking a leader vehicle in both simulations and real-world flight tests. Our results demonstrate that maximizing observability contributed to improving global positioning precision for the quadrotor team., Comment: 12 pages, 5 figures

Programming a quantum device describes the usage of quantum logic gates, agnostic of hardware specifics, to perform a sequence of operations with (typically) a computing or sensing task in mind. Such programs have been executed on digital quantum computers, which despite their noisy character, have shown the ability to optimize metrological functions, for example in the generation of spin squeezing and optimization of quantum Fisher information for signals manifesting as spin rotations in a quantum register. However, the qubits of these programmable quantum sensors are tightly spatially confined and therefore suboptimal for enclosing the kinds of large spacetime areas required for performing inertial sensing. In this work, we derive a set of quantum logic gates for a cold atom optical lattice interferometer that manipulates the momentum of atoms. Here, the operations are framed in terms of single qubit operations and mappings between qubit subspaces with internal levels given by the Bloch (crystal) eigenstates of the lattice. We describe how the quantum optimal control method of direct collocation is well suited for obtaining the modulation waveforms of the lattice which achieve these operations.

One dimensional fluid/electron Monte Carlo simulations of capacitively coupled Ar/O2 discharges driven by sawtooth up voltage waveforms are performed as a function of the number of consecutive harmonics driving frequencies of 13.56 MHz, N (1-3), pressure (200-500 mTorr) and gas mixture (10-90 % admixture of O2 to Ar). The effects of these external parameters on the electron dynamics, and the transport of ions and neutrals are revealed at constant peak-to-peak driving voltage. The electronegativity is found to decline as the number of consecutive harmonics increases and the DC self-bias voltage decreases. Increasing the pressure also leads to a decrease in electronegativity. The combination of a decrease in the mean free path of electrons and the presence of the Electrical Asymmetry Effect (EAE) result in different spatio-temporal distributions of the ionization rate, which lead to a reduction in the amplitude of the DC self-bias at higher pressure. As the admixture of electronegative O2 increases, the electronegativity is enhanced, and the discharge mode changes from an {\alpha}-Drift Ambipolar (DA) hybrid to DA mode. This work focuses on linking these fundamental changes of the plasma physics induced by changing external parameters to process relevant charged particle and neutral fluxes to the electrodes. Particular attention is paid to O(1D) flux, because it is a precursor of deposition. In discharges driven by sawtooth up voltage waveforms, placing the substrate on the grounded electrode and increasing the number of consecutive harmonics, N, can facilitate the deposition process, since the O(1D) flux to the substrate is higher in these scenarios. Moreover, at an O2 admixture of 20%, the O(1D) flux is nearly as high as that at an O2 admixture of 90%, indicating that a higher O(1D) flux can be achieved without excessively increasing the O2 admixture., Comment: Ar/O2 gas discharges, electron dynamics, transport of charged and neutral particles, sawtooth up voltage waveforms

Diffeomorphic image registration is crucial for various medical imaging applications because it can preserve the topology of the transformation. This study introduces DCCNN-LSTM-Reg, a learning framework that evolves dynamically and learns a symmetrical registration path by satisfying a specified control increment system. This framework aims to obtain symmetric diffeomorphic deformations between moving and fixed images. To achieve this, we combine deep learning networks with diffeomorphic mathematical mechanisms to create a continuous and dynamic registration architecture, which consists of multiple Symmetric Registration (SR) modules cascaded on five different scales. Specifically, our method first uses two U-nets with shared parameters to extract multiscale feature pyramids from the images. We then develop an SR-module comprising a sequential CNN-LSTM architecture to progressively correct the forward and reverse multiscale deformation fields using control increment learning and the homotopy continuation technique. Through extensive experiments on three 3D registration tasks, we demonstrate that our method outperforms existing approaches in both quantitative and qualitative evaluations., Comment: 12 pages,7 figures

Multi-objective Bayesian optimization has been widely adopted in scientific experiment design, including drug discovery and hyperparameter optimization. In practice, regulatory or safety concerns often impose additional thresholds on certain attributes of the experimental outcomes. Previous work has primarily focused on constrained single-objective optimization tasks or active search under constraints. We propose CMOBO, a sample-efficient constrained multi-objective Bayesian optimization algorithm that balances learning of the feasible region (defined on multiple unknowns) with multi-objective optimization within the feasible region in a principled manner. We provide both theoretical justification and empirical evidence, demonstrating the efficacy of our approach on various synthetic benchmarks and real-world applications.

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

In this paper we present a novel method for efficient and effective 3D surface reconstruction in open scenes. Existing Neural Radiance Fields (NeRF) based works typically require extensive training and rendering time due to the adopted implicit representations. In contrast, 3D Gaussian splatting (3DGS) uses an explicit and discrete representation, hence the reconstructed surface is built by the huge number of Gaussian primitives, which leads to excessive memory consumption and rough surface details in sparse Gaussian areas. To address these issues, we propose Gaussian Voxel Kernel Functions (GVKF), which establish a continuous scene representation based on discrete 3DGS through kernel regression. The GVKF integrates fast 3DGS rasterization and highly effective scene implicit representations, achieving high-fidelity open scene surface reconstruction. Experiments on challenging scene datasets demonstrate the efficiency and effectiveness of our proposed GVKF, featuring with high reconstruction quality, real-time rendering speed, significant savings in storage and training memory consumption., Comment: NeurIPS 2024

Localization is one of the most crucial tasks for Unmanned Aerial Vehicle systems (UAVs) directly impacting overall performance, which can be achieved with various sensors and applied to numerous tasks related to search and rescue operations, object tracking, construction, etc. However, due to the negative effects of challenging environments, UAVs may lose signals for localization. In this paper, we present an effective path-planning system leveraging semantic segmentation information to navigate around texture-less and problematic areas like lakes, oceans, and high-rise buildings using a monocular camera. We introduce a real-time semantic segmentation architecture and a novel keyframe decision pipeline to optimize image inputs based on pixel distribution, reducing processing time. A hierarchical planner based on the Dynamic Window Approach (DWA) algorithm, integrated with a cost map, is designed to facilitate efficient path planning. The system is implemented in a photo-realistic simulation environment using Unity, aligning with segmentation model parameters. Comprehensive qualitative and quantitative evaluations validate the effectiveness of our approach, showing significant improvements in the reliability and efficiency of UAV localization in challenging environments., Comment: In The 24th International Conference on Control, Automation, and Systems (ICCAS 2024), Jeju, Korea

Navigating safely in dynamic human environments is crucial for mobile service robots, and social navigation is a key aspect of this process. In this paper, we proposed an integrative approach that combines motion prediction and trajectory planning to enable safe and socially-aware robot navigation. The main idea of the proposed method is to leverage the advantages of Socially Acceptable trajectory prediction and Timed Elastic Band (TEB) by incorporating human interactive information including position, orientation, and motion into the objective function of the TEB algorithms. In addition, we designed social constraints to ensure the safety of robot navigation. The proposed system is evaluated through physical simulation using both quantitative and qualitative metrics, demonstrating its superior performance in avoiding human and dynamic obstacles, thereby ensuring safe navigation. The implementations are open source at: \url{https://github.com/thanhnguyencanh/SGan-TEB.git}, Comment: In the 24th International Conference on Control, Automation, and Systems (ICCAS 2024), Jeju, Korea

Background: Cone-beam computed tomography (CBCT) plays a crucial role in image-guided radiotherapy, but artifacts and noise make them unsuitable for accurate dose calculation. Artificial intelligence methods have shown promise in enhancing CBCT quality to produce synthetic CT (sCT) images. However, existing methods either produce images of suboptimal quality or incur excessive time costs, failing to satisfy clinical practice standards. Methods and materials: We propose a novel hybrid conditional latent diffusion model for efficient and accurate CBCT-to-CT synthesis, named HC$^3$L-Diff. We employ the Unified Feature Encoder (UFE) to compress images into a low-dimensional latent space, thereby optimizing computational efficiency. Beyond the use of CBCT images, we propose integrating its high-frequency knowledge as a hybrid condition to guide the diffusion model in generating sCT images with preserved structural details. This high-frequency information is captured using our designed High-Frequency Extractor (HFE). During inference, we utilize denoising diffusion implicit model to facilitate rapid sampling. We construct a new in-house prostate dataset with paired CBCT and CT to validate the effectiveness of our method. Result: Extensive experimental results demonstrate that our approach outperforms state-of-the-art methods in terms of sCT quality and generation efficiency. Moreover, our medical physicist conducts the dosimetric evaluations to validate the benefit of our method in practical dose calculation, achieving a remarkable 93.8% gamma passing rate with a 2%/2mm criterion, superior to other methods. Conclusion: The proposed HC$^3$L-Diff can efficiently achieve high-quality CBCT-to-CT synthesis in only over 2 mins per patient. Its promising performance in dose calculation shows great potential for enhancing real-world adaptive radiotherapy., Comment: 13 pages, 5 figures

The multi-modal posterior under unidentified nonparametric models yields poor mixing of Markov Chain Monte Carlo (MCMC), which is a stumbling block to Bayesian predictions. In this article, we conceptualize a prior informativeness threshold that is essentially the variance of posterior modes and expressed by the uncertainty hyperparameters of nonparametric priors. The threshold plays the role of a lower bound of the within-chain MCMC variance to ensure MCMC mixing, and engines prior modification through hyperparameter tuning to descend the mode variance. Our method distinguishes from existing postprocessing methods in that it directly samples well-mixed MCMC chains on the unconstrained space, and inherits the original posterior predictive distribution in predictive inference. Our method succeeds in Bayesian survival predictions under an unidentified nonparametric transformation model, guarded by the inferential theories of the posterior variance, under elicitation of two delicate nonparametric priors. Comprehensive simulations and real-world data analysis demonstrate that our method achieves MCMC mixing and outperforms existing approaches in survival predictions.

Flat bands typically describe energy bands whose energy dispersion is entirely or almost entirely degenerate. One effective method to form flat bands is by constructing Moir\'e superlattices. Recently, there has been a shift in perspective regarding the roles of space (momentum) and time (energy) in a lattice, with the concept of photonic time crystals that has sparked discussions on momentum dispersion such as the presence of a bandgap in momentum. Here we propose a photonic time moir\'e superlattice achieved by overlaying two photonic time crystals with different periods. The resulting momentum bandgap of this superlattice supports isolated momentum bands that are nearly independent of energy, which we refer to as momentum flat bands. Unlike energy flat bands, which have zero group velocity, momentum flat bands exhibit infinitely large group velocity across a broad frequency range. Unlike previous optical media supporting broadband superluminal propagation based on gain, the effective refractive index of the momentum flat bands is real-valued, leading to more stabilized superluminal pulse propagation., Comment: 6 pages, 4 figures

We consider the problem of physically-based inverse rendering using 3D Gaussian Splatting (3DGS) representations. While recent 3DGS methods have achieved remarkable results in novel view synthesis (NVS), accurately capturing high-fidelity geometry, physically interpretable materials and lighting remains challenging, as it requires precise geometry modeling to provide accurate surface normals, along with physically-based rendering (PBR) techniques to ensure correct material and lighting disentanglement. Previous 3DGS methods resort to approximating surface normals, but often struggle with noisy local geometry, leading to inaccurate normal estimation and suboptimal material-lighting decomposition. In this paper, we introduce GeoSplatting, a novel hybrid representation that augments 3DGS with explicit geometric guidance and differentiable PBR equations. Specifically, we bridge isosurface and 3DGS together, where we first extract isosurface mesh from a scalar field, then convert it into 3DGS points and formulate PBR equations for them in a fully differentiable manner. In GeoSplatting, 3DGS is grounded on the mesh geometry, enabling precise surface normal modeling, which facilitates the use of PBR frameworks for material decomposition. This approach further maintains the efficiency and quality of NVS from 3DGS while ensuring accurate geometry from the isosurface. Comprehensive evaluations across diverse datasets demonstrate the superiority of GeoSplatting, consistently outperforming existing methods both quantitatively and qualitatively., Comment: Project page: https://pku-vcl-geometry.github.io/GeoSplatting/

The magnetised near detector (ND280) of the T2K long-baseline neutrino oscillation experiment has been recently upgraded aiming to satisfy the requirement of reducing the systematic uncertainty from measuring the neutrinonucleus interaction cross section, which is the largest systematic uncertainty in the search for leptonic charge-parity symmetry violation. A key component of the upgrade is SuperFGD, a 3D segmented plastic scintillator detector made of approximately 2,000,000 optically-isolated 1 cm3 cubes. It will provide a 3D image of GeV neutrino interactions by combining tracking and stopping power measurements of final state particles with sub-nanosecond time resolution. The performance of SuperFGD is characterized by the precision of its response to charged particles as well as the systematic effects that might affect the physics measurements. Hence, a detailed Geant4 based optical simulation of the SuperFGD building block, i.e. a plastic scintillating cube read out by three wavelength shifting fibers, has been developed and validated with the different datasets collected in various beam tests. In this manuscript the description of the optical model as well as the comparison with data are reported., Comment: 31 pages, 15 figures

Recent advances in quantum information processing with superconducting qubits have fueled a growing demand for scaling and miniaturizing circuit layouts. Despite significant progress, accurately predicting the Hamiltonian of complex circuits remains a challenging task. In this work, we propose an improved method for quantizing superconducting circuits that incorporates material- and geometry-dependent kinetic inductance. Our approach models thin superconducting films as equivalent reactive boundary elements, seamlessly integrating into the conventional circuit quantization framework without adding computational complexity. As a proof of concept, we experimentally verify our method using planar superconducting quantum devices made of 35 nm-thick disordered niobium films, known to exhibit large kinetic inductance values. We demonstrate significantly improved accuracy in predicting the Hamiltonian based solely on the chip layout and material properties of the superconducting films and Josephson junctions, with discrepancies in mode frequencies remaining below 2%. Our method enables systematic studies of superconducting devices based on disordered thin films or compact, fine-pitched elements and, more broadly, facilitates the precise engineering of superconducting quantum circuits at scale., Comment: 12 pages, 4 figures, 1 table

Motivated by the study of a variant of sunflowers, Alon and Holzman recently introduced focal-free hypergraphs. In this paper, we show that there is an interesting connection between the maximum size of focal-free hypergraphs and the renowned Erd\H{o}s Matching Conjecture on the maximum number of edges that can be contained in a uniform hypergraph with bounded matching number. As a consequence, we give asymptotically optimal bounds on the maximum sizes of focal-free uniform hypergraphs and codes, thereby significantly improving the previous results of Alon and Holzman. Moreover, by using the existentce results of combinatorial designs and orthogonal arrays, we are able to explicitly determine the exact sizes of maximum focal-free uniform hypergraphs and codes for a wide range of parameters.

Terahertz (THz) integrated sensing and communication (ISAC) technology is envisioned to achieve high communication performance alongside advanced sensing abilities. For various applications of ISAC, accurate environment reconstruction including geometry reconstruction and material identification is critical. This paper presents a highly precise geometry reconstruction algorithm and material identification scheme for a monostatic sensing case in a typical indoor scenario. Experiments are conducted in the frequency range from 290 GHz to 310 GHz using a vector network analyzer (VNA)-based channel sounder by co-locating the transmitter and receiver. A joint delay and angle space-alternating generalized expectation-maximization (SAGE)-based algorithm is implemented to estimate multipath component (MPC) parameters and the indoor geometry is reconstructed based on the extracted parameters. Furthermore, a geometry-based method is employed to model and remove the spurious path of the corner, reaching an accuracy of 1.75 cm. Additionally, a material database using THz time-domain spectroscopy (THz-TDS) is established, capturing reflection losses of over 200 common material samples. Applying this database to our monostatic sensing, the measured reflection losses of wall and window frame are accurately identified as cement and steel, respectively. Our results demonstrate the centimeter-level geometry reconstruction and accurate material identification for practical THz ISAC scenarios, which unleash unprecedented sensing potential compared to microwave and millimeter-wave bands.

We consider the problem of novel view synthesis from unposed images in a single feed-forward. Our framework capitalizes on fast speed, scalability, and high-quality 3D reconstruction and view synthesis capabilities of 3DGS, where we further extend it to offer a practical solution that relaxes common assumptions such as dense image views, accurate camera poses, and substantial image overlaps. We achieve this through identifying and addressing unique challenges arising from the use of pixel-aligned 3DGS: misaligned 3D Gaussians across different views induce noisy or sparse gradients that destabilize training and hinder convergence, especially when above assumptions are not met. To mitigate this, we employ pre-trained monocular depth estimation and visual correspondence models to achieve coarse alignments of 3D Gaussians. We then introduce lightweight, learnable modules to refine depth and pose estimates from the coarse alignments, improving the quality of 3D reconstruction and novel view synthesis. Furthermore, the refined estimates are leveraged to estimate geometry confidence scores, which assess the reliability of 3D Gaussian centers and condition the prediction of Gaussian parameters accordingly. Extensive evaluations on large-scale real-world datasets demonstrate that PF3plat sets a new state-of-the-art across all benchmarks, supported by comprehensive ablation studies validating our design choices., Comment: project page: https://cvlab-kaist.github.io/PF3plat/

This paper proposes an unsupervised deep-learning (DL) approach by integrating transformer and Kolmogorov-Arnold networks (KAN) termed KANsformer to realize scalable beamforming for mobile communication systems. Specifically, we consider a classic multi-input-single-output energy efficiency maximization problem subject to the total power budget. The proposed KANsformer first extracts hidden features via a multi-head self-attention mechanism and then reads out the desired beamforming design via KAN. Numerical results are provided to evaluate the KANsformer in terms of generalization performance, transfer learning and ablation experiment. Overall, the KANsformer outperforms existing benchmark DL approaches, and is adaptable to the change in the number of mobile users with real-time and near-optimal inference.

GPT-4o is an autoregressive omni model that accepts as input any combination of text, audio, image, and video, and generates any combination of text, audio, and image outputs. It's trained end-to-end across text, vision, and audio, meaning all inputs and outputs are processed by the same neural network. GPT-4o can respond to audio inputs in as little as 232 milliseconds, with an average of 320 milliseconds, which is similar to human response time in conversation. It matches GPT-4 Turbo performance on text in English and code, with significant improvement on text in non-English languages, while also being much faster and 50\% cheaper in the API. GPT-4o is especially better at vision and audio understanding compared to existing models. In line with our commitment to building AI safely and consistent with our voluntary commitments to the White House, we are sharing the GPT-4o System Card, which includes our Preparedness Framework evaluations. In this System Card, we provide a detailed look at GPT-4o's capabilities, limitations, and safety evaluations across multiple categories, focusing on speech-to-speech while also evaluating text and image capabilities, and measures we've implemented to ensure the model is safe and aligned. We also include third-party assessments on dangerous capabilities, as well as discussion of potential societal impacts of GPT-4o's text and vision capabilities.

Two-dimensional (2D) hybrid organic-inorganic perovskite (HOIP) demonstates great potential for developing flexible and wearable spintronic devices, by serving as spin sources via the bulk Rashba effect (BRE). However, the practical application of BRE in 2D HOIP faces huge challenges, particularly due to the toxicity of lead, which is crucial for achieving large spin-orbit coupling, and the restrictions in 2D HOIP candidates to meet specific symmetry-breaking requirements. To overcome these obstacles, we design a strategy to exploit the interfacial Rashba effect (IRE) of lead-free 2D HOIP (C6H5CH2CH2NH3)2CuCl4 (PEA-CuCl), manifesting as an efficient spin generator at room temperature. IRE of PEA-CuCl originates from the large orbital hybridization at the interface between PEA-CuCl and adjacent ferromagnetic layers. Spin-torque ferromagnetic resonance measurements further quantify a large Rashba effective field of 14.04 Oe per 10^11 A m-2, surpassing those of lead-based HOIP and traditional all-inorganic heterojunctions with noble metals. Our lead-free 2D HOIP PEA-CuCl, which harnesses large IRE for spin generation, is efficient, nontoxic, and economic, offering huge promise for future flexible and wearable spintronic devices.

Full-duplex spoken dialogue systems significantly advance over traditional turn-based dialogue systems, as they allow simultaneous bidirectional communication, closely mirroring human-human interactions. However, achieving low latency and natural interactions in full-duplex dialogue systems remains a significant challenge, especially considering human conversation dynamics such as interruptions, backchannels, and overlapping speech. In this paper, we introduce a novel End-to-End GPT-based model OmniFlatten for full-duplex conversation, capable of effectively modeling the complex behaviors inherent to natural conversations with low latency. To achieve full-duplex communication capabilities, we propose a multi-stage post-training scheme that progressively adapts a text-based large language model (LLM) backbone into a speech-text dialogue LLM, capable of generating text and speech in real time, without modifying the architecture of the backbone LLM. The training process comprises three stages: modality alignment, half-duplex dialogue learning, and full-duplex dialogue learning. Throughout all training stages, we standardize the data using a flattening operation, which allows us to unify the training methods and the model architecture across different modalities and tasks. Our approach offers a straightforward modeling technique and a promising research direction for developing efficient and natural end-to-end full-duplex spoken dialogue systems. Audio samples of dialogues generated by OmniFlatten can be found at this web site (https://omniflatten.github.io/)., Comment: Work in progress

We present a comprehensive study of spontaneous vectorization in the Einstein-Maxwell-Vector (EMV) model, uncovering a novel vectorization mechanism driven by the competition between electromagnetic and vector fields. By utilizing a generalized coordinate transformation, we resolve previous divergences near the event horizon and significantly expand the domain of existence for vectorized Reissner-Nordstr\"om black holes (VRNBHs). Our results highlight the crucial role of a newly defined combined charge $ \sqrt{\tilde{Q}^2 + \tilde{P}^2} $, which encapsulates this competition and establishes a clear thermodynamic preference, and a more natural description of light rights for VRNBHs over standard Reissner-Nordstr\"om black holes. Moreover, we identify unique features in the light ring structure of VRNBHs, revealing deeper connections to multi-charge black holes and opening new avenues for observational signatures in effective field theory., Comment: 31 pages, 11 figures

In this paper, we introduce Janus, an autoregressive framework that unifies multimodal understanding and generation. Prior research often relies on a single visual encoder for both tasks, such as Chameleon. However, due to the differing levels of information granularity required by multimodal understanding and generation, this approach can lead to suboptimal performance, particularly in multimodal understanding. To address this issue, we decouple visual encoding into separate pathways, while still leveraging a single, unified transformer architecture for processing. The decoupling not only alleviates the conflict between the visual encoder's roles in understanding and generation, but also enhances the framework's flexibility. For instance, both the multimodal understanding and generation components can independently select their most suitable encoding methods. Experiments show that Janus surpasses previous unified model and matches or exceeds the performance of task-specific models. The simplicity, high flexibility, and effectiveness of Janus make it a strong candidate for next-generation unified multimodal models., Comment: Technical Report

Researchers have recently achieved significant advances in deep learning techniques, which in turn has substantially advanced other research disciplines, such as natural language processing, image processing, speech recognition, and software engineering. Various deep learning techniques have been successfully employed to facilitate software engineering tasks, including code generation, software refactoring, and fault localization. Many papers have also been presented in top conferences and journals, demonstrating the applications of deep learning techniques in resolving various software engineering tasks. However, although several surveys have provided overall pictures of the application of deep learning techniques in software engineering, they focus more on learning techniques, that is, what kind of deep learning techniques are employed and how deep models are trained or fine-tuned for software engineering tasks. We still lack surveys explaining the advances of subareas in software engineering driven by deep learning techniques, as well as challenges and opportunities in each subarea. To this end, in this paper, we present the first task-oriented survey on deep learning-based software engineering. It covers twelve major software engineering subareas significantly impacted by deep learning techniques. Such subareas spread out the through the whole lifecycle of software development and maintenance, including requirements engineering, software development, testing, maintenance, and developer collaboration. As we believe that deep learning may provide an opportunity to revolutionize the whole discipline of software engineering, providing one survey covering as many subareas as possible in software engineering can help future research push forward the frontier of deep learning-based software engineering more systematically., Comment: Accepted in SCIENCE CHINA Information Sciences

Vision Language Models (VLMs) often struggle with culture-specific knowledge, particularly in languages other than English and in underrepresented cultural contexts. To evaluate their understanding of such knowledge, we introduce WorldCuisines, a massive-scale benchmark for multilingual and multicultural, visually grounded language understanding. This benchmark includes a visual question answering (VQA) dataset with text-image pairs across 30 languages and dialects, spanning 9 language families and featuring over 1 million data points, making it the largest multicultural VQA benchmark to date. It includes tasks for identifying dish names and their origins. We provide evaluation datasets in two sizes (12k and 60k instances) alongside a training dataset (1 million instances). Our findings show that while VLMs perform better with correct location context, they struggle with adversarial contexts and predicting specific regional cuisines and languages. To support future research, we release a knowledge base with annotated food entries and images along with the VQA data., Comment: Preprint

This paper introduces a novel approach that integrates future closest point predictions into the distance constraints of a collision avoidance controller, leveraging convex hulls with closest point distance calculations. By addressing abrupt shifts in closest points, this method effectively reduces collision risks and enhances controller performance. Applied to an Image Guided Therapy robot and validated through simulations and user experiments, the framework demonstrates improved distance prediction accuracy, smoother trajectories, and safer navigation near obstacles.

We present a spatial and angular Gaussian based representation and a triple splatting process, for real-time, high-quality novel lighting-and-view synthesis from multi-view point-lit input images. To describe complex appearance, we employ a Lambertian plus a mixture of angular Gaussians as an effective reflectance function for each spatial Gaussian. To generate self-shadow, we splat all spatial Gaussians towards the light source to obtain shadow values, which are further refined by a small multi-layer perceptron. To compensate for other effects like global illumination, another network is trained to compute and add a per-spatial-Gaussian RGB tuple. The effectiveness of our representation is demonstrated on 30 samples with a wide variation in geometry (from solid to fluffy) and appearance (from translucent to anisotropic), as well as using different forms of input data, including rendered images of synthetic/reconstructed objects, photographs captured with a handheld camera and a flash, or from a professional lightstage. We achieve a training time of 40-70 minutes and a rendering speed of 90 fps on a single commodity GPU. Our results compare favorably with state-of-the-art techniques in terms of quality/performance. Our code and data are publicly available at https://GSrelight.github.io/., Comment: Accepted to SIGGRAPH Asia 2024. Project page: https://gsrelight.github.io/

Reduction of detailed chemical reaction mechanisms is one of the key methods for mitigating the computational cost of reactive flow simulations. Exploitation of species and elementary reaction sparsity ensures the compactness of the reduced mechanisms. In this work, we propose a novel sparse statistical learning approach for chemical reaction mechanism reduction. Specifically, the reduced mechanism is learned to explicitly reproduce the dynamical evolution of detailed chemical kinetics, while constraining on the sparsity of the reduced reactions at the same time. Compact reduced mechanisms are be achieved as the collection of species that participate in the identified important reactions. We validate our approach by reducing oxidation mechanisms for $n$-heptane (194 species) and 1,3-butadiene (581 species). The results demonstrate that the reduced mechanisms show accurate predictions for the ignition delay times, laminar flame speeds, species mole fraction profiles and turbulence-chemistry interactions across a wide range of operating conditions. Comparative analysis with directed relation graph (DRG)-based methods and the state-of-the-art (SOTA) methods reveals that our sparse learning approach produces reduced mechanisms with fewer species while maintaining the same error limits. The advantages are particularly evident for detailed mechanisms with a larger number of species and reactions. The sparse learning strategy shows significant potential in achieving more substantial reductions in complex chemical reaction mechanisms.

Person re-identification(ReID), as a crucial technology in the field of security, plays a vital role in safety inspections, personnel counting, and more. Most current ReID approaches primarily extract features from images, which are easily affected by objective conditions such as clothing changes and occlusions. In addition to cameras, we leverage widely available routers as sensing devices by capturing gait information from pedestrians through the Channel State Information (CSI) in WiFi signals and contribute a multimodal dataset. We employ a two-stream network to separately process video understanding and signal analysis tasks, and conduct multi-modal fusion and contrastive learning on pedestrian video and WiFi data. Extensive experiments in real-world scenarios demonstrate that our method effectively uncovers the correlations between heterogeneous data, bridges the gap between visual and signal modalities, significantly expands the sensing range, and improves ReID accuracy across multiple sensors.

In response to climate change and urban heat island effects, enhancing human thermal comfort in cities is crucial for sustainable urban development. Traditional methods for investigating the urban thermal environment and corresponding human thermal comfort level are often resource intensive, inefficient, and limited in scope. To address these challenges, we (1) introduce the concept of thermal affordance, which represents the inherent capacity of a streetscape to influence human thermal comfort based on its visual and physical features; and (2) a method to evaluate it (visual assessment of thermal affordance -- VATA), which combines street view imagery (SVI), online and in-filed surveys, and statistical learning algorithms. VATA extracts five categories of image features from SVI data and establishes 19 visual-perceptual indicators for streetscape visual assessment. Using a multi-task neural network and elastic net regression, we model their chained relationship to predict and comprehend thermal affordance for Singapore. VATA predictions are validated with field-investigated OTC data, providing a cost-effective and scalable method to assess the thermal comfort potential of urban streetscape. This framework can inform streetscape design to support sustainable, livable, and resilient urban environments.

Achieving a balance between accuracy and efficiency is a critical challenge in facial landmark detection (FLD). This paper introduces the Parallel Optimal Position Search (POPoS), a high-precision encoding-decoding framework designed to address the fundamental limitations of traditional FLD methods. POPoS employs three key innovations: (1) Pseudo-range multilateration is utilized to correct heatmap errors, enhancing the precision of landmark localization. By integrating multiple anchor points, this approach minimizes the impact of individual heatmap inaccuracies, leading to robust overall positioning. (2) To improve the pseudo-range accuracy of selected anchor points, a new loss function, named multilateration anchor loss, is proposed. This loss function effectively enhances the accuracy of the distance map, mitigates the risk of local optima, and ensures optimal solutions. (3) A single-step parallel computation algorithm is introduced, significantly enhancing computational efficiency and reducing processing time. Comprehensive evaluations across five benchmark datasets demonstrate that POPoS consistently outperforms existing methods, particularly excelling in low-resolution scenarios with minimal computational overhead. These features establish POPoS as a highly efficient and accurate tool for FLD, with broad applicability in real-world scenarios. The code is available at https://github.com/teslatasy/PoPoS, Comment: The experimental setup and results require further modifications and improvements to ensure the accuracy and quality of the paper. Therefore, we are requesting to withdraw the submission

Enlarging the context window of large language models (LLMs) has become a crucial research area, particularly for applications involving extremely long texts. In this work, we propose a novel training-free framework for processing long texts, utilizing a divide-and-conquer strategy to achieve comprehensive document understanding. The proposed LLM$\times$MapReduce framework splits the entire document into several chunks for LLMs to read and then aggregates the intermediate answers to produce the final output. The main challenge for divide-and-conquer long text processing frameworks lies in the risk of losing essential long-range information when splitting the document, which can lead the model to produce incomplete or incorrect answers based on the segmented texts. Disrupted long-range information can be classified into two categories: inter-chunk dependency and inter-chunk conflict. We design a structured information protocol to better cope with inter-chunk dependency and an in-context confidence calibration mechanism to resolve inter-chunk conflicts. Experimental results demonstrate that LLM$\times$MapReduce can outperform representative open-source and commercial long-context LLMs, and is applicable to several different models., Comment: Work in Progress. Code: https://github.com/thunlp/LLMxMapReduce

LLM-powered coding and development assistants have become prevalent to programmers' workflows. However, concerns about the trustworthiness of LLMs for code persist despite their widespread use. Much of the existing research focused on either training or evaluation, raising questions about whether stakeholders in training and evaluation align in their understanding of model trustworthiness and whether they can move toward a unified direction. In this paper, we propose a vision for a unified trustworthiness auditing framework, DataTrust, which adopts a data-centric approach that synergistically emphasizes both training and evaluation data and their correlations. DataTrust aims to connect model trustworthiness indicators in evaluation with data quality indicators in training. It autonomously inspects training data and evaluates model trustworthiness using synthesized data, attributing potential causes from specific evaluation data to corresponding training data and refining indicator connections. Additionally, a trustworthiness arena powered by DataTrust will engage crowdsourced input and deliver quantitative outcomes. We outline the benefits that various stakeholders can gain from DataTrust and discuss the challenges and opportunities it presents., Comment: Short Vision Paper

Code Large Language Models (LLMs) have demonstrated remarkable capabilities in generating, understanding, and manipulating programming code. However, their training process inadvertently leads to the memorization of sensitive information, posing severe privacy risks. Existing studies on memorization in LLMs primarily rely on prompt engineering techniques, which suffer from limitations such as widespread hallucination and inefficient extraction of the target sensitive information. In this paper, we present a novel approach to characterize real and fake secrets generated by Code LLMs based on token probabilities. We identify four key characteristics that differentiate genuine secrets from hallucinated ones, providing insights into distinguishing real and fake secrets. To overcome the limitations of existing works, we propose DESEC, a two-stage method that leverages token-level features derived from the identified characteristics to guide the token decoding process. DESEC consists of constructing an offline token scoring model using a proxy Code LLM and employing the scoring model to guide the decoding process by reassigning token likelihoods. Through extensive experiments on four state-of-the-art Code LLMs using a diverse dataset, we demonstrate the superior performance of DESEC in achieving a higher plausible rate and extracting more real secrets compared to existing baselines. Our findings highlight the effectiveness of our token-level approach in enabling an extensive assessment of the privacy leakage risks associated with Code LLMs., Comment: 12 pages, 8 figures

Significant success has been achieved in fabricating angstrom-scale artificial solid ionic channels aiming to replicate the biological ion channels (BICs).Besides high selectivity, BICs also exhibit sophisticated ion gating and interplay. However, such behavior and functionality are seldomly recreated in the artificial counterparts due to the insufficient understanding of the molecular origin. Here we report cooperative and inhibitory ion transport in angstrom-scale acetate functionalized MoS2 two dimensional channels., Comment: 34 pages for main text and 40 pages for SI

Exceptional points (EPs) are singularities in the parameter space of a non-Hermitian system where eigenenergies and eigenstates coincide. They hold promise for enhancing sensing applications, but this is limited by the divergence of shot noise near EPs. According to recent studies, EP sensors operating in the nonlinear regime may avoid these limitations. By analyzing an exemplary nonlinear system, we show that the interplay of noise and nonlinearity introduces previously-unidentified obstacles to enhanced sensing. The noise effectively displaces the EP in parameter space and reduces its order, thereby eliminating the sought-for divergence in the signal-to-noise ratio. Moreover, the noise near the nonlinear EP experiences a stronger divergence than predicted by standard calculations of the Petermann noise factor, due to the properties of the Bogoliubov-de Gennes Hamiltonian governing the fluctuations. Our semi-analytical estimates for the noise level agree quantitatively with the results of stochastic numerical simulations.

Large Language Models (LLMs) are susceptible to security and safety threats, such as prompt injection, prompt extraction, and harmful requests. One major cause of these vulnerabilities is the lack of an instruction hierarchy. Modern LLM architectures treat all inputs equally, failing to distinguish between and prioritize various types of instructions, such as system messages, user prompts, and data. As a result, lower-priority user prompts may override more critical system instructions, including safety protocols. Existing approaches to achieving instruction hierarchy, such as delimiters and instruction-based training, do not address this issue at the architectural level. We introduce the Instructional Segment Embedding (ISE) technique, inspired by BERT, to modern large language models, which embeds instruction priority information directly into the model. This approach enables models to explicitly differentiate and prioritize various instruction types, significantly improving safety against malicious prompts that attempt to override priority rules. Our experiments on the Structured Query and Instruction Hierarchy benchmarks demonstrate an average robust accuracy increase of up to 15.75% and 18.68%, respectively. Furthermore, we observe an improvement in instruction-following capability of up to 4.1% evaluated on AlpacaEval. Overall, our approach offers a promising direction for enhancing the safety and effectiveness of LLM architectures., Comment: Preprint

Recently, differentiable mask pruning methods optimize the continuous relaxation architecture (soft network) as the proxy of the pruned discrete network (hard network) for superior sub-architecture search. However, due to the agnostic impact of the discretization process, the hard network struggles with the equivalent representational capacity as the soft network, namely discretization gap, which severely spoils the pruning performance. In this paper, we first investigate the discretization gap and propose a novel structural differentiable mask pruning framework named S2HPruner to bridge the discretization gap in a one-stage manner. In the training procedure, SH2Pruner forwards both the soft network and its corresponding hard network, then distills the hard network under the supervision of the soft network. To optimize the mask and prevent performance degradation, we propose a decoupled bidirectional knowledge distillation. It blocks the weight updating from the hard to the soft network while maintaining the gradient corresponding to the mask. Compared with existing pruning arts, S2HPruner achieves surpassing pruning performance without fine-tuning on comprehensive benchmarks, including CIFAR-100, Tiny ImageNet, and ImageNet with a variety of network architectures. Besides, investigation and analysis experiments explain the effectiveness of S2HPruner. Codes will be released soon., Comment: NeurIPS 2024 accepted

Consider a network of agents that all want to guess the correct value of some ground truth state. In a sequential order, each agent makes its decision using a single private signal which has a constant probability of error, as well as observations of actions from its network neighbors earlier in the order. We are interested in enabling \emph{network-wide asymptotic truth learning} -- that in a network of $n$ agents, almost all agents make a correct prediction with probability approaching one as $n$ goes to infinity. In this paper we study both random orderings and carefully crafted decision orders with respect to the graph topology as well as sufficient or necessary conditions for a graph to support such a good ordering. We first show that on a sparse graph of average constant degree with a random ordering asymptotic truth learning does not happen. We then show a rather modest sufficient condition to enable asymptotic truth learning. With the help of this condition we characterize graphs generated from the Erd\"os R\'enyi model and preferential attachment model. In an Erd\"os R\'enyi graph, unless the graph is super sparse (with $O(n)$ edges) or super dense (nearly a complete graph), there exists a decision ordering that supports asymptotic truth learning. Similarly, any preferential attachment network with a constant number of edges per node can achieve asymptotic truth learning under a carefully designed ordering but not under either a random ordering nor the arrival order. We also evaluated a variant of the decision ordering on different network topologies and demonstrated clear effectiveness in improving truth learning over random orderings., Comment: Accepted at the 20th Conference on Web and Internet Economics (WINE'24)

This paper studies robust forward investment and consumption preferences and optimal strategies for a risk-averse and ambiguity-averse agent in an incomplete financial market with drift and volatility uncertainties. We focus on non-zero volatility and constant relative risk aversion (CRRA) forward preferences. Given the non-convexity of the Hamiltonian with respect to uncertain volatilities, we first construct robust randomized forward preferences through endogenous randomization in an auxiliary market. We derive the corresponding optimal and robust investment and consumption strategies. Furthermore, we show that such forward preferences and strategies, developed in the auxiliary market, remain optimal and robust in the physical market, offering a comprehensive framework for forward investment and consumption under model uncertainty., Comment: 36 pages

Augmenting the multi-step reasoning abilities of Large Language Models (LLMs) has been a persistent challenge. Recently, verification has shown promise in improving solution consistency by evaluating generated outputs. However, current verification approaches suffer from sampling inefficiencies, requiring a large number of samples to achieve satisfactory performance. Additionally, training an effective verifier often depends on extensive process supervision, which is costly to acquire. In this paper, we address these limitations by introducing a novel verification method based on Twisted Sequential Monte Carlo (TSMC). TSMC sequentially refines its sampling effort to focus exploration on promising candidates, resulting in more efficient generation of high-quality solutions. We apply TSMC to LLMs by estimating the expected future rewards at partial solutions. This approach results in a more straightforward training target that eliminates the need for step-wise human annotations. We empirically demonstrate the advantages of our method across multiple math benchmarks, and also validate our theoretical analysis of both our approach and existing verification methods.

In this paper, we investigate the performance of the cross-domain iterative detection (CDID) framework with orthogonal time frequency space (OTFS) modulation, where two distinct CDID algorithms are presented. The proposed schemes estimate/detect the information symbols iteratively across the frequency domain and the delay-Doppler (DD) domain via passing either the a posteriori or extrinsic information. Building upon this framework, we investigate the error performance by considering the bias evolution and state evolution. Furthermore, we discuss their error performance in convergence and the DD domain error state lower bounds in each iteration. Specifically, we demonstrate that in convergence, the ultimate error performance of the CDID passing the a posteriori information can be characterized by two potential convergence points. In contrast, the ultimate error performance of the CDID passing the extrinsic information has only one convergence point, which, interestingly, aligns with the matched filter bound. Our numerical results confirm our analytical findings and unveil the promising error performance achieved by the proposed designs.