Search

In this paper, we introduce ProtoOcc, a novel 3D occupancy prediction model designed to predict the occupancy states and semantic classes of 3D voxels through a deep semantic understanding of scenes. ProtoOcc consists of two main components: the Dual Branch Encoder (DBE) and the Prototype Query Decoder (PQD). The DBE produces a new 3D voxel representation by combining 3D voxel and BEV representations across multiple scales through a dual branch structure. This design enhances both performance and computational efficiency by providing a large receptive field for the BEV representation while maintaining a smaller receptive field for the voxel representation. The PQD introduces Prototype Queries to accelerate the decoding process. Scene-Adaptive Prototypes are derived from the 3D voxel features of input sample, while Scene-Agnostic Prototypes are computed by applying Scene-Adaptive Prototypes to an Exponential Moving Average during the training phase. By using these prototype-based queries for decoding, we can directly predict 3D occupancy in a single step, eliminating the need for iterative Transformer decoding. Additionally, we propose the Robust Prototype Learning, which injects noise into prototype generation process and trains the model to denoise during the training phase. ProtoOcc achieves state-of-the-art performance with 45.02% mIoU on the Occ3D-nuScenes benchmark. For single-frame method, it reaches 39.56% mIoU with an inference speed of 12.83 FPS on an NVIDIA RTX 3090. Our code can be found at https://github.com/SPA-junghokim/ProtoOcc., Comment: Accepted to AAAI Conference on Artificial Intelligence 2025, 9 pages, 5 figures

This technical report introduces the EXAONE 3.5 instruction-tuned language models, developed and released by LG AI Research. The EXAONE 3.5 language models are offered in three configurations: 32B, 7.8B, and 2.4B. These models feature several standout capabilities: 1) exceptional instruction following capabilities in real-world scenarios, achieving the highest scores across seven benchmarks, 2) outstanding long-context comprehension, attaining the top performance in four benchmarks, and 3) competitive results compared to state-of-the-art open models of similar sizes across nine general benchmarks. The EXAONE 3.5 language models are open to anyone for research purposes and can be downloaded from https://huggingface.co/LGAI-EXAONE. For commercial use, please reach out to the official contact point of LG AI Research: contact_us@lgresearch.ai., Comment: arXiv admin note: text overlap with arXiv:2408.03541

We present the first circularly polarized Floquet engineering time-resolved Resonant Inelastic X-ray Scattering (tr-RIXS) experiment in H$_3$LiIr$_2$O$_6$, an iridium-based Kitaev system. Our calculations and experimental results are consistent with the modification of the low energy magnetic excitations in H$_3$LiIr$_2$O$_6$ only during illumination by the laser pulse, consistent with the Floquet engineering of the exchange interactions. However, the penetration length mismatch between the X-ray probe and laser pump and the intrinsic complexity of Kitaev magnets prevented us from unequivocally extracting towards which ground H$_3$LiIr$_2$O$_6$ was driven. We outline possible solutions to these challenges for Floquet stabilization and observation of the Kitaev Quantum Spin Liquid limit by RIXS.

Achieving robust stereo 3D imaging under diverse illumination conditions is an important however challenging task, due to the limited dynamic ranges (DRs) of cameras, which are significantly smaller than real world DR. As a result, the accuracy of existing stereo depth estimation methods is often compromised by under- or over-exposed images. Here, we introduce dual-exposure stereo for extended dynamic range 3D imaging. We develop automatic dual-exposure control method that adjusts the dual exposures, diverging them when the scene DR exceeds the camera DR, thereby providing information about broader DR. From the captured dual-exposure stereo images, we estimate depth using motion-aware dual-exposure stereo network. To validate our method, we develop a robot-vision system, collect stereo video datasets, and generate a synthetic dataset. Our method outperforms other exposure control methods.

We present BootComp, a novel framework based on text-to-image diffusion models for controllable human image generation with multiple reference garments. Here, the main bottleneck is data acquisition for training: collecting a large-scale dataset of high-quality reference garment images per human subject is quite challenging, i.e., ideally, one needs to manually gather every single garment photograph worn by each human. To address this, we propose a data generation pipeline to construct a large synthetic dataset, consisting of human and multiple-garment pairs, by introducing a model to extract any reference garment images from each human image. To ensure data quality, we also propose a filtering strategy to remove undesirable generated data based on measuring perceptual similarities between the garment presented in human image and extracted garment. Finally, by utilizing the constructed synthetic dataset, we train a diffusion model having two parallel denoising paths that use multiple garment images as conditions to generate human images while preserving their fine-grained details. We further show the wide-applicability of our framework by adapting it to different types of reference-based generation in the fashion domain, including virtual try-on, and controllable human image generation with other conditions, e.g., pose, face, etc., Comment: Project page: https://yisol.github.io/BootComp

We report measurements of production cross sections for $\rho^+$, $\rho^0$, $\omega$, $K^{*+}$, $K^{*0}$, $\phi$, $\eta$, $K_S^0$, $f_0(980)$, $D^+$, $D^0$, $D_s^+$, $D^{*+}$, $D^{*0}$, and $D^{*+}_s$ in $e^+e^-$ collisions at a center-of-mass energy near 10.58 GeV. The data were recorded by the Belle experiment, consisting of 571 fb$^{-1}$ at 10.58 GeV and 74 fb$^{-1}$ at 10.52 GeV. Production cross sections are extracted as a function of the fractional hadron momentum $x_p$ . The measurements are compared to {\sc pythia} Monte Carlo generator predictions with various fragmentation settings, including those that have increased fragmentation into vector mesons over pseudo-scalar mesons. The cross sections measured for light hadrons are consistent with no additional increase of vector over pseudo-scalar mesons. The charmed-meson cross sections are compared to earlier measurements -- when available -- including older Belle results, which they supersede. They are in agreement before application of an improved initial-state radiation correction procedure that causes slight changes in their \xp shapes., Comment: 21 pages, 18 figures, submitted to Phys. Rev. D

We present 850 $\mu$m polarization observations of the IC 348 star-forming region in the Perseus molecular cloud as part of the B-fields In STar-forming Region Observation (BISTRO) survey. We study the magnetic properties of two cores (HH 211 MMS and IC 348 MMS) and a filamentary structure of IC 348. We find that the overall field tends to be more perpendicular than parallel to the filamentary structure of the region. The polarization fraction decreases with intensity, and we estimate the trend by power-law and the mean of the Rice distribution fittings. The power indices for the cores are much smaller than 1, indicative of possible grain growth to micron size in the cores. We also measure the magnetic field strengths of the two cores and the filamentary area separately by applying the Davis-Chandrasekhar-Fermi method and its alternative version for compressed medium. The estimated mass-to-flux ratios are 0.45-2.20 and 0.63-2.76 for HH 211 MMS and IC 348 MMS, respectively, while the ratios for the filament is 0.33-1.50. This result may suggest that the transition from subcritical to supercritical conditions occurs at the core scale ($\sim$ 0.05 pc) in the region. In addition, we study the energy balance of the cores and find that the relative strength of turbulence to the magnetic field tends to be stronger for IC 348 MMS than HH 211 MMS. The result could potentially explain the different configurations inside the two cores: a single protostellar system in HH 211 MMS and multiple protostars in IC 348 MMS., Comment: Accepted for publication in ApJ. 21 pages, 12 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

In the bottomonium sector, the hindered magnetic dipole (M1) transitions between P-wave states $h_b(2P) \rightarrow \chi_{bJ}(1P) \gamma$, $J=0, \, 1, \, 2$, are expected to be severely suppressed according to the Relativized Quark Model, due to the spin flip of the $b$ quark. Nevertheless, a recent model following the coupled-channel approach predicts the corresponding branching fractions to be enhanced by orders of magnitude. In this Letter, we report the first search for such transitions. We find no significant signals and set upper limits at 90% CL on the corresponding branching fractions: $\mathcal{B}[h_b(2P)\to\gamma\chi_{b0}(1P)] < 2.7 \times 10^{-1}$, $\mathcal{B}[h_b(2P)\to\gamma\chi_{b1}(1P)] < 5.4 \times 10^{-3}$ and $\mathcal{B}[h_b(2P)\to\gamma\chi_{b2}(1P)] < 1.3 \times 10^{-2}$. These values help to constrain the parameters of the coupled-channel models. The results are obtained using a $121.4 \, fb^{-1}$ data sample taken around $\sqrt{s}= 10.860 \, GeV$ with the Belle detector at the KEKB asymmetric-energy $e^+e^-$ collider.

Accelerating end-to-end inference of transformer-based large language models (LLMs) is a critical component of AI services in datacenters. However, diverse compute characteristics of end-to-end LLM inference present challenges as previously proposed accelerators only address certain operations or stages (e.g., self-attention, generation stage, etc.). To address the unique challenges of accelerating end-to-end inference, we propose IANUS -- Integrated Accelerator based on NPU-PIM Unified Memory System. IANUS is a domain-specific system architecture that combines a Neural Processing Unit (NPU) with a Processing-in-Memory (PIM) to leverage both the NPU's high computation throughput and the PIM's high effective memory bandwidth. In particular, IANUS employs a unified main memory system where the PIM memory is used both for PIM operations and for NPU's main memory. The unified main memory system ensures that memory capacity is efficiently utilized and the movement of shared data between NPU and PIM is minimized. However, it introduces new challenges since normal memory accesses and PIM computations cannot be performed simultaneously. Thus, we propose novel PIM Access Scheduling that manages normal memory accesses and PIM computations through workload mapping and scheduling across the PIM and the NPU. Our detailed simulation evaluations show that IANUS improves the performance of GPT-2 by 6.2$\times$ and 3.2$\times$, on average, compared to the NVIDIA A100 GPU and the state-of-the-art accelerator. As a proof-of-concept, we develop a prototype of IANUS with a commercial PIM, NPU, and an FPGA-based PIM controller to demonstrate the feasibility of IANUS., Comment: Updated version of the paper accepted to ASPLOS 2024

We examine diverging preferences in human-labeled preference datasets. We develop a taxonomy of disagreement sources spanning 10 categories across four high-level classes -- task underspecification, response style, refusals, and annotation errors. We find that the majority of disagreements are in opposition with standard reward modeling approaches, which are designed with the assumption that annotator disagreement is noise. We then explore how these findings impact two areas of LLM development: reward modeling and evaluation. In our experiments, we demonstrate how standard reward modeling methods, like the Bradley-Terry model, fail to differentiate whether a given preference judgment is the result of unanimous agreement among annotators or the majority opinion among diverging user preferences. We also find that these tendencies are also echoed by popular LLM-as-Judge evaluation methods, which consistently identify a winning response in cases of diverging preferences. These findings highlight remaining challenges in LLM evaluations, which are greatly influenced by divisive features like response style, and in developing pluralistically aligned LLMs. To address these issues, we develop methods for identifying diverging preferences to mitigate their influence on evaluation and training.

The goal of this paper is to accelerate codec-based speech synthesis systems with minimum sacrifice to speech quality. We propose an enhanced inference method that allows for flexible trade-offs between speed and quality during inference without requiring additional training. Our core idea is to predict multiple tokens per inference step of the AR module using multiple prediction heads, resulting in a linear reduction in synthesis time as the number of heads increases. Furthermore, we introduce a novel speculative decoding technique that utilises a Viterbi-based algorithm to select the optimal sequence of generated tokens at each decoding step. In our experiments, we demonstrate that the time required to predict each token is reduced by a factor of 4 to 5 compared to baseline models, with minimal quality trade-off or even improvement in terms of speech intelligibility. Audio samples are available at: multpletokensprediction.github.io/multipletokensprediction.github.io/., Comment: Submitted to IEEE ICASSP 2025

Large language models (LLMs) have demonstrated remarkable performance in the legal domain, with GPT-4 even passing the Uniform Bar Exam in the U.S. However their efficacy remains limited for non-standardized tasks and tasks in languages other than English. This underscores the need for careful evaluation of LLMs within each legal system before application. Here, we introduce KBL, a benchmark for assessing the Korean legal language understanding of LLMs, consisting of (1) 7 legal knowledge tasks (510 examples), (2) 4 legal reasoning tasks (288 examples), and (3) the Korean bar exam (4 domains, 53 tasks, 2,510 examples). First two datasets were developed in close collaboration with lawyers to evaluate LLMs in practical scenarios in a certified manner. Furthermore, considering legal practitioners' frequent use of extensive legal documents for research, we assess LLMs in both a closed book setting, where they rely solely on internal knowledge, and a retrieval-augmented generation (RAG) setting, using a corpus of Korean statutes and precedents. The results indicate substantial room and opportunities for improvement., Comment: EMNLP 2024 Findings

Large Language Models (LLMs) excel at understanding the semantic relationships between queries and documents, even with lengthy and complex long-tail queries. These queries are challenging for feedback-based rankings due to sparse user engagement and limited feedback, making LLMs' ranking ability highly valuable. However, the large size and slow inference of LLMs necessitate the development of smaller, more efficient models (sLLMs). Recently, integrating ranking label generation into distillation techniques has become crucial, but existing methods underutilize LLMs' capabilities and are cumbersome. Our research, RRADistill: Re-Ranking Ability Distillation, propose an efficient label generation pipeline and novel sLLM training methods for both encoder and decoder models. We introduce an encoder-based method using a Term Control Layer to capture term matching signals and a decoder-based model with a ranking layer for enhanced understanding. A/B testing on a Korean-based search platform, validates the effectiveness of our approach in improving re-ranking for long-tail queries., Comment: Accepted to EMNLP 2024 Industry Track. First two authors contributed equally

Large language models (LLMs) have achieved significant performance gains using advanced prompting techniques over various tasks. However, the increasing length of prompts leads to high computational costs and often obscures crucial information. Prompt compression has been proposed to alleviate these issues, but it faces challenges in (i) capturing the global context and (ii) training the compressor effectively. To tackle these challenges, we introduce a novel prompt compression method, namely Reading To Compressing (R2C), utilizing the Fusion-in-Decoder (FiD) architecture to identify the important information in the prompt. Specifically, the cross-attention scores of the FiD are used to discern essential chunks and sentences from the prompt. R2C effectively captures the global context without compromising semantic consistency while detouring the necessity of pseudo-labels for training the compressor. Empirical results show that R2C retains key contexts, enhancing the LLM performance by 6% in out-of-domain evaluations while reducing the prompt length by 80%., Comment: Findings of the Association for Computational Linguistics: EMNLP 2024; 21 pages; 10 figures and 7 tables

A common challenge in aggregating data from multiple sources can be formalized as an \textit{Optimal Transport} (OT) barycenter problem, which seeks to compute the average of probability distributions with respect to OT discrepancies. However, the presence of outliers and noise in the data measures can significantly hinder the performance of traditional statistical methods for estimating OT barycenters. To address this issue, we propose a novel, scalable approach for estimating the \textit{robust} continuous barycenter, leveraging the dual formulation of the \textit{(semi-)unbalanced} OT problem. To the best of our knowledge, this paper is the first attempt to develop an algorithm for robust barycenters under the continuous distribution setup. Our method is framed as a $\min$-$\max$ optimization problem and is adaptable to \textit{general} cost function. We rigorously establish the theoretical underpinnings of the proposed method and demonstrate its robustness to outliers and class imbalance through a number of illustrative experiments., Comment: 19 pages, 4 figures

Optimal Transport (OT) theory investigates the cost-minimizing transport map that moves a source distribution to a target distribution. Recently, several approaches have emerged for learning the optimal transport map for a given cost function using neural networks. We refer to these approaches as the OT Map. OT Map provides a powerful tool for diverse machine learning tasks, such as generative modeling and unpaired image-to-image translation. However, existing methods that utilize max-min optimization often experience training instability and sensitivity to hyperparameters. In this paper, we propose a novel method to improve stability and achieve a better approximation of the OT Map by exploiting displacement interpolation, dubbed Displacement Interpolation Optimal Transport Model (DIOTM). We derive the dual formulation of displacement interpolation at specific time $t$ and prove how these dual problems are related across time. This result allows us to utilize the entire trajectory of displacement interpolation in learning the OT Map. Our method improves the training stability and achieves superior results in estimating optimal transport maps. We demonstrate that DIOTM outperforms existing OT-based models on image-to-image translation tasks., Comment: 20 pages

Unpaired point cloud completion explores methods for learning a completion map from unpaired incomplete and complete point cloud data. In this paper, we propose a novel approach for unpaired point cloud completion using the unbalanced optimal transport map, called Unbalanced Optimal Transport Map for Unpaired Point Cloud Completion (UOT-UPC). We demonstrate that the unpaired point cloud completion can be naturally interpreted as the Optimal Transport (OT) problem and introduce the Unbalanced Optimal Transport (UOT) approach to address the class imbalance problem, which is prevalent in unpaired point cloud completion datasets. Moreover, we analyze the appropriate cost function for unpaired completion tasks. This analysis shows that the InfoCD cost function is particularly well-suited for this task. Our model is the first attempt to leverage UOT for unpaired point cloud completion, achieving competitive or superior results on both single-category and multi-category datasets. In particular, our model is especially effective in scenarios with class imbalance, where the proportions of categories are different between the incomplete and complete point cloud datasets., Comment: 20 pages, 10 figures

The Optimal Transport (OT) problem investigates a transport map that connects two distributions while minimizing a given cost function. Finding such a transport map has diverse applications in machine learning, such as generative modeling and image-to-image translation. In this paper, we introduce a scalable and simulation-free approach for solving the Entropic Unbalanced Optimal Transport (EUOT) problem. We derive the dynamical form of this EUOT problem, which is a generalization of the Schr\"odinger bridges (SB) problem. Based on this, we derive dual formulation and optimality conditions of the EUOT problem from the stochastic optimal control interpretation. By leveraging these properties, we propose a simulation-free algorithm to solve EUOT, called Simulation-free EUOT (SF-EUOT). While existing SB models require expensive simulation costs during training and evaluation, our model achieves simulation-free training and one-step generation by utilizing the reciprocal property. Our model demonstrates significantly improved scalability in generative modeling and image-to-image translation tasks compared to previous SB methods., Comment: 26 pages

Place recognition plays an important role in achieving robust long-term autonomy. Real-world robots face a wide range of weather conditions (e.g. overcast, heavy rain, and snowing) and most sensors (i.e. camera, LiDAR) essentially functioning within or near-visible electromagnetic waves are sensitive to adverse weather conditions, making reliable localization difficult. In contrast, radar is gaining traction due to long electromagnetic waves, which are less affected by environmental changes and weather independence. In this work, we propose a radar-based lightweight and robust place recognition. We achieve rotational invariance and lightweight by selecting a one-dimensional ring-shaped description and robustness by mitigating the impact of false detection utilizing opposite noise characteristics between free space and feature. In addition, the initial heading can be estimated, which can assist in building a SLAM pipeline that combines odometry and registration, which takes into account onboard computing. The proposed method was tested for rigorous validation across various scenarios (i.e. single session, multi-session, and different weather conditions). In particular, we validate our descriptor achieving reliable place recognition performance through the results of extreme environments that lacked structural information such as an OORD dataset., Comment: 8 pages, 8 figures, accepted to RA-L

Assessment refers to a broad array of approaches for measuring or evaluating a person's (or group of persons') skills, behaviors, dispositions, or other attributes. Assessments range from standardized tests used in admissions, employee selection, licensure examinations, and domestic and international largescale assessments of cognitive and behavioral skills to formative K-12 classroom curricular assessments. The various types of assessments are used for a wide variety of purposes, but they also have many common elements, such as standards for their reliability, validity, and fairness--even classroom assessments have standards (Klinger et al., 2015). The authors believe the future of assessments will involve a shift in emphasis on what skills will be measured, innovations in how one goes about measuring them, the use of advanced technologies for test operations, and an expansion in the value and kinds of information that test-takers will receive from taking the assessment. In this paper, the authors argue and provide evidence for the belief that the future of assessment contains challenges but is promising. The challenges include risks associated with security and exposure of personal data, test score bias, and inappropriate test uses, all of which may be exacerbated by the growing infiltration of artificial intelligence (AI) into people's lives. The promise is increasing opportunities for testing to help individuals achieve their education and career goals and contribute to well-being and overall quality of life. To help achieve this promise the authors focus on the evidence-based science of measurement in education and workplace learning, a theme throughout this paper. The first section of the paper reviews evidence for the value of assessment and discusses how the role of assessment may expand as skills become the new currency. The many purposes of assessment are discussed, from high-stakes examinations and selection tests to low-stakes formative assessments. The emerging challenges to testing and assessment, related to perceptions about their value, their focus, validity, fairness and equity concerns, are reviewed. The authors conclude the first section with a discussion of the prospects for the future of assessments, including the capacity of assessment to provide useful information to test-takers, the importance of identifying key skills and advancing methods for assessing hard-to-measure skills, and the importance of providing opportunities with personalized feedback. The remaining sections of the paper address those themes. The intended audience for this report is broad-ranging from the international scientific community in areas engaged in assessment, particularly education and workforce, to policy-makers and funders in those areas. The authors try to strike a balance between technical detail and accessibility to the broad audience.

We present a microscopic theory of DC current-biased Josephson junctions applicable to arbitrary junction transparencies, resolving long debates on subharmonic gap structure (SGS). Our approach extends beyond the existing theories relying on phenomenological models or the lowest order perturbation valid for opaque junctions. In discord with experiments, those theories fail to obtain all singularities of SGS at voltages equalling integer fractions of twice the superconducting gap. We address this limitation using a microscopic Floquet approach applicable to junctions with arbitrary transparencies, capturing all subharmonic singularities. We explain the origin of the subharmonics using two complementary pictures, in time domain exploiting the interference of non-equilibrium current pulses, and in frequency domain using a generalisation of multiple Andreev reflections., Comment: 8 pages, 5 figures

Programmatically generated synthetic data has been used in differential private training for classification to enhance performance without privacy leakage. However, as the synthetic data is generated from a random process, the distribution of real data and the synthetic data are distinguishable and difficult to transfer. Therefore, the model trained with the synthetic data generates unrealistic random images, raising challenges to adapt the synthetic data for generative models. In this work, we propose DP-SynGen, which leverages programmatically generated synthetic data in diffusion models to address this challenge. By exploiting the three stages of diffusion models(coarse, context, and cleaning) we identify stages where synthetic data can be effectively utilized. We theoretically and empirically verified that cleaning and coarse stages can be trained without private data, replacing them with synthetic data to reduce the privacy budget. The experimental results show that DP-SynGen improves the quality of generative data by mitigating the negative impact of privacy-induced noise on the generation process.

Gyrokinetic simulations are utilized to study effects of magnetic islands on the ion temperature gradient (ITG) turbulence in the KSTAR tokamak with resonant magnetic perturbations. Simulations show that the transport is controlled by the nonlinear interactions between the ITG turbulence and self-generated vortex flows and zonal flows, leading to an anisotropic structure of fluctuation and transport on the poloidal plane and in the toroidal direction. Magnetic islands greatly enhance turbulent transport of both particle and heat. The turbulent transport exhibits variations in the toroidal direction, with transport through the resonant layer near the island X-point being enhanced when the X-point is located at the outer mid-plane. A quantitative agreement is shown between simulations and KSTAR experiments in terms of time frequency and perpendicular wavevector spectrum.

Large-scale eigenvalue problems arise in various fields of science and engineering and demand computationally efficient solutions. In this study, we investigate the subspace approximation for parametric linear eigenvalue problems, aiming to mitigate the computational burden associated with high-fidelity systems. We provide general error estimates under non-simple eigenvalue conditions, establishing the theoretical foundations for our methodology. Numerical examples, ranging from one-dimensional to three-dimensional setups, are presented to demonstrate the efficacy of reduced basis method in handling parametric variations in boundary conditions and coefficient fields to achieve significant computational savings while maintaining high accuracy, making them promising tools for practical applications in large-scale eigenvalue computations.

We report on neutron diffraction, magnetoresistance, magnetization, and magnetic torque measurements under high magnetic field in the helical antiferromagnet CeVGe$_3$. This compound exhibits Kondo lattice coherence and helical antiferromagnetic (AFM) ordering at ambient pressure, similar to the well-studied CeRhIn$_5$. Our measurements reveal that CeVGe$_3$ undergoes a magnetic transition from an incommensurate (ICM) AFM state to an up-up-down-down commensurate (CM) AFM structure, followed by a transition to a novel phase at higher fields. A quantum phase transition occurs around 21.3 T. This rich magnetic field phase diagram closely resembles that of CeRhIn$_5$. Furthermore, angle-dependent magnetoresistance measurements reveal that all transitions in CeVGe$_3$ occur from the field component along the $ab$ plane. These findings highlight the intricate interplay among exchange interactions, crystal field effects, ground state properties, and crystalline symmetries., Comment: 8 pages, 9 figures, accepted by Physics Review B

Articulated object manipulation is a challenging task, requiring constrained motion and adaptive control to handle the unknown dynamics of the manipulated objects. While reinforcement learning (RL) has been widely employed to tackle various scenarios and types of articulated objects, the complexity of these tasks, stemming from multiple intertwined objectives makes learning a control policy in the full task space highly difficult. To address this issue, we propose a Subspace-wise hybrid RL (SwRL) framework that learns policies for each divided task space, or subspace, based on independent objectives. This approach enables adaptive force modulation to accommodate the unknown dynamics of objects. Additionally, it effectively leverages the previously underlooked redundant subspace, thereby maximizing the robot's dexterity. Our method enhances both learning efficiency and task execution performance, as validated through simulations and real-world experiments. Supplementary video is available at https://youtu.be/PkNxv0P8Atk, Comment: 14 pages, 10 figures, Submitted to Robotics and Autonomous Systems

Let $\mathrm{Int}(n)$ denote the set of nonempty left weak Bruhat intervals in the symmetric group $\mathfrak{S}_n$. We investigate the equivalence relation $\overset{D}{\simeq}$ on $\mathrm{Int}(n)$, where $I \overset{D}{\simeq} J$ if and only if there exists a descent-preserving poset isomorphism between $I$ and $J$. For each equivalence class $C$ of $(\mathrm{Int}(n), \overset{D}{\simeq})$, a partial order $\preceq$ is defined by $[\sigma, \rho]_L \preceq [\sigma', \rho']_L$ if and only if $\sigma \preceq_R \sigma'$. Kim-Lee-Oh (2023) showed that the poset $(C, \preceq)$ is isomorphic to a right weak Bruhat interval. In this paper, we focus on lower and upper descent weak Bruhat intervals, specifically those of the form $[w_0(S), \sigma]_L$ or $[\sigma, w_1(S)]_L$, where $w_0(S)$ is the longest element in the parabolic subgroup $\mathfrak{S}_S$ of $\mathfrak{S}_n$, generated by $\{s_i \mid i \in S\}$ for a subset $S \subseteq [n-1]$, and $w_1(S)$ is the longest element among the minimal-length representatives of left $\mathfrak{S}_{[n-1] \setminus S}$-cosets in $\mathfrak{S}_n$. We begin by providing a poset-theoretic characterization of the equivalence relation $\overset{D}{\simeq}$. Using this characterization, the minimal and maximal elements within an equivalence class $C$ are identified when $C$ is a lower or upper descent interval. Under an additional condition, a detailed description of the structure of $(C, \preceq)$ is provided. Furthermore, for the equivalence class containing $[w_0(S), \sigma]_L$, an injective hull of ${\sf B}([w_0(S), \sigma]_L)$ is given, and for the equivalence class containing $[\sigma, w_1(S)]_L$, a projective cover of ${\sf B}([\sigma, w_1(S)]_L)$ is given., Comment: 48 pages

Cosmic shear, galaxy clustering, and the abundance of massive halos each probe the large-scale structure of the universe in complementary ways. We present cosmological constraints from the joint analysis of the three probes, building on the latest analyses of the lensing-informed abundance of clusters identified by the South Pole Telescope (SPT) and of the auto- and cross-correlation of galaxy position and weak lensing measurements (3$\times$2pt) in the Dark Energy Survey (DES). We consider the cosmological correlation between the different tracers and we account for the systematic uncertainties that are shared between the large-scale lensing correlation functions and the small-scale lensing-based cluster mass calibration. Marginalized over the remaining $\Lambda$CDM parameters (including the sum of neutrino masses) and 52 astrophysical modeling parameters, we measure $\Omega_\mathrm{m}=0.300\pm0.017$ and $\sigma_8=0.797\pm0.026$. Compared to constraints from Planck primary CMB anisotropies, our constraints are only 15% wider with a probability to exceed of 0.22 ($1.2\sigma$) for the two-parameter difference. We further obtain $S_8\equiv\sigma_8(\Omega_\mathrm{m}/0.3)^{0.5}=0.796\pm0.013$ which is lower than the Planck measurement at the $1.6\sigma$ level. The combined SPT cluster, DES 3$\times$2pt, and Planck datasets mildly prefer a non-zero positive neutrino mass, with a 95% upper limit $\sum m_\nu<0.25~\mathrm{eV}$ on the sum of neutrino masses. Assuming a $w$CDM model, we constrain the dark energy equation of state parameter $w=-1.15^{+0.23}_{-0.17}$ and when combining with Planck primary CMB anisotropies, we recover $w=-1.20^{+0.15}_{-0.09}$, a $1.7\sigma$ difference with a cosmological constant. The precision of our results highlights the benefits of multiwavelength multiprobe cosmology., Comment: Submitted to Phys. Rev. D

In sequential recommendation (SR), neural models have been actively explored due to their remarkable performance, but they suffer from inefficiency inherent to their complexity. On the other hand, linear SR models exhibit high efficiency and achieve competitive or superior accuracy compared to neural models. However, they solely deal with the sequential order of items (i.e., sequential information) and overlook the actual timestamp (i.e., temporal information). It is limited to effectively capturing various user preference drifts over time. To address this issue, we propose a novel linear SR model, named TemporAl LinEar item-item model (TALE), incorporating temporal information while preserving training/inference efficiency, with three key components. (i) Single-target augmentation concentrates on a single target item, enabling us to learn the temporal correlation for the target item. (ii) Time interval-aware weighting utilizes the actual timestamp to discern the item correlation depending on time intervals. (iii) Trend-aware normalization reflects the dynamic shift of item popularity over time. Our empirical studies show that TALE outperforms ten competing SR models by up to 18.71% gains on five benchmark datasets. It also exhibits remarkable effectiveness in evaluating long-tail items by up to 30.45% gains. The source code is available at https://github.com/psm1206/TALE., Comment: Accepted by WSDM 2025

Large Language Models (LLMs) have recently demonstrated significant potential in the field of time series forecasting, offering impressive capabilities in handling complex temporal data. However, their robustness and reliability in real-world applications remain under-explored, particularly concerning their susceptibility to adversarial attacks. In this paper, we introduce a targeted adversarial attack framework for LLM-based time series forecasting. By employing both gradient-free and black-box optimization methods, we generate minimal yet highly effective perturbations that significantly degrade the forecasting accuracy across multiple datasets and LLM architectures. Our experiments, which include models like TimeGPT and LLM-Time with GPT-3.5, GPT-4, LLaMa, and Mistral, show that adversarial attacks lead to much more severe performance degradation than random noise, and demonstrate the broad effectiveness of our attacks across different LLMs. The results underscore the critical vulnerabilities of LLMs in time series forecasting, highlighting the need for robust defense mechanisms to ensure their reliable deployment in practical applications., Comment: 11 pages, 5 figures

The ability to reconstruct high-fidelity fluid flow fields from sparse sensor measurement is critical for many science and engineering applications, but remains a huge challenge. This challenge is caused by the large difference between the dimensions of the state space and the observational space, making the operator that provides the mapping from the state space to the observational space ill-conditioned and non-invertible. As a result, deriving the forward map from the observational space to the state space as the inverse of the measurement operator is nearly impossible. While traditional methods, including sparse optimization, data assimilation, or machine learning based regressive reconstruction, are available, they often struggle with generalization and computational efficiency, particularly when non-uniform or varying discretization of the domain are considered. In this work, we propose FLRONet, a novel operator learning framework designed to reconstruct full-state flow fields from sparse sensor measurements in space and time. FLRONet utilizes a branch-trunk architecture, where the branch network integrates sensor observations from multiple time instances, and the trunk network encodes the entire temporal domain. This design allows FLRONet to achieve highly accurate, discretization-independent reconstructions at any time within the observation window. Although the popular three-dimensional Fourier Neural Operator offers similar functionalities, our results show that FLRONet surpasses it in both accuracy and efficiency. FLRONet not only achieves superior performance in approximating the true operator but also exhibits significantly faster inference at high-fidelity discretizations.

Timely detection and treatment are essential for maintaining eye health. Visual acuity (VA), which measures the clarity of vision at a distance, is a crucial metric for managing eye health. Machine learning (ML) techniques have been introduced to assist in VA measurement, potentially alleviating clinicians' workloads. However, the inherent uncertainties in ML models make relying solely on them for VA prediction less than ideal. The VA prediction task involves multiple sources of uncertainty, requiring more robust approaches. A promising method is to build prediction sets or intervals rather than point estimates, offering coverage guarantees through techniques like conformal prediction and Probably Approximately Correct (PAC) prediction sets. Despite the potential, to date, these approaches have not been applied to the VA prediction task.To address this, we propose a method for deriving prediction intervals for estimating visual acuity from fundus images with a PAC guarantee. Our experimental results demonstrate that the PAC guarantees are upheld, with performance comparable to or better than that of two prior works that do not provide such guarantees., Comment: To be published in ML4H 2024

Using $(27.12\pm 0.14)\times 10^{8}$ $\psi(3686)$ events collected with the BESIII detector operating at the BEPCII collider, the decay $\psi(3686)\to\Sigma^{0}\bar{\Sigma}^{0}\phi$ is observed for the first time with a statistical significance of 7.6$\sigma$. Its branching fraction is measured to be $(2.64 \pm 0.32_{\textrm{stat}} \pm 0.12_{\textrm{sys}}) \times 10^{-6}$, where the first uncertainty is statistical and the second is systematic. In addition, we search for potential intermediate states in the $\Sigma^{0}\phi$($\bar{\Sigma}^{0}\phi$) invariant mass distribution and a possible threshold enhancement in the $\Sigma^{0}\bar{\Sigma}^{0}$ system, but no conclusive evidence of is observed.

We articulate fundamental mismatches between technical methods for machine unlearning in Generative AI, and documented aspirations for broader impact that these methods could have for law and policy. These aspirations are both numerous and varied, motivated by issues that pertain to privacy, copyright, safety, and more. For example, unlearning is often invoked as a solution for removing the effects of targeted information from a generative-AI model's parameters, e.g., a particular individual's personal data or in-copyright expression of Spiderman that was included in the model's training data. Unlearning is also proposed as a way to prevent a model from generating targeted types of information in its outputs, e.g., generations that closely resemble a particular individual's data or reflect the concept of "Spiderman." Both of these goals--the targeted removal of information from a model and the targeted suppression of information from a model's outputs--present various technical and substantive challenges. We provide a framework for thinking rigorously about these challenges, which enables us to be clear about why unlearning is not a general-purpose solution for circumscribing generative-AI model behavior in service of broader positive impact. We aim for conceptual clarity and to encourage more thoughtful communication among machine learning (ML), law, and policy experts who seek to develop and apply technical methods for compliance with policy objectives., Comment: Presented at the 2nd Workshop on Generative AI and Law at ICML (July 2024)

Using a sample of $(2712\pm14)\times10^6$ $\psi(3686)$ events collected with the BESIII detector, we perform partial wave analyses of the decays $\psi(3686)\to p\bar{p}\pi^0$ and $\psi(3686)\to p\bar{p}\eta$. The branching fractions of $\psi(3686)\to p\bar{p}\pi^0$ and $\psi(3686)\to p\bar{p}\eta$ are determined to be $(133.9\pm11.2\pm2.3)\times10^{-6}$ or $(183.7\pm13.7\pm3.2)\times10^{-6}$ and $(61.5\pm6.5\pm1.1)\times10^{-6}$ or $(84.4\pm6.9\pm1.4)\times10^{-6}$, respectively, where the two solutions are caused by an ambiguous phase angle between resonant and continuum processes. Several well-established $N^*$ states are observed in the $p\pi^0$ and $p\eta$ systems, and the corresponding branching fractions are measured. The ratio of decay widths $\Gamma_{N(1535)\to N\eta}/\Gamma_{N(1535)\to N\pi}$ is determined to be $0.99\pm0.05\pm0.17$., Comment: 19 pages, 7 figures, submitted to PRD

Maritime environments often present hazardous situations due to factors such as moving ships or buoys, which become obstacles under the influence of waves. In such challenging conditions, the ability to detect and track potentially hazardous objects is critical for the safe navigation of marine robots. To address the scarcity of comprehensive datasets capturing these dynamic scenarios, we introduce a new multi-modal dataset that includes image and point-wise annotations of maritime hazards. Our dataset provides detailed ground truth for obstacle detection and tracking, including objects as small as 10$\times$10 pixels, which are crucial for maritime safety. To validate the dataset's effectiveness as a reliable benchmark, we conducted evaluations using various methodologies, including \ac{SOTA} techniques for object detection and tracking. These evaluations are expected to contribute to performance improvements, particularly in the complex maritime environment. To the best of our knowledge, this is the first dataset offering multi-modal annotations specifically tailored to maritime environments. Our dataset is available at https://sites.google.com/view/polaris-dataset.

Beam Drift Chamber (BDC) is designed to reconstruct the trajectories of incident rare isotope beams provided by RAON (Rare isotope Accelerator complex for ON-line experiments) into the experimental target of LAMPS (Large Acceptance Multi-Purpose Spectrometer). To conduct the performance test of the BDC, the prototype BDC (pBDC) is manufactured and evaluated with the high energy ion beams from HIMAC (Heavy Ion Medical Accelerator in Chiba) facility in Japan. Two kinds of ion beams, 100 MeV proton, and 200 MeV/u $^{12}$C, have been utilized for this evaluation, and the track reconstruction efficiency and position resolution have been measured as the function of applied high voltage. This paper introduces the construction details and presents the track reconstruction efficiency and position resolution of pBDC., Comment: 13 pages, 15 figures

Circuits based on sum-product structure have become a ubiquitous representation to compactly encode knowledge, from Boolean functions to probability distributions. By imposing constraints on the structure of such circuits, certain inference queries become tractable, such as model counting and most probable configuration. Recent works have explored analyzing probabilistic and causal inference queries as compositions of basic operators to derive tractability conditions. In this paper, we take an algebraic perspective for compositional inference, and show that a large class of queries - including marginal MAP, probabilistic answer set programming inference, and causal backdoor adjustment - correspond to a combination of basic operators over semirings: aggregation, product, and elementwise mapping. Using this framework, we uncover simple and general sufficient conditions for tractable composition of these operators, in terms of circuit properties (e.g., marginal determinism, compatibility) and conditions on the elementwise mappings. Applying our analysis, we derive novel tractability conditions for many such compositional queries. Our results unify tractability conditions for existing problems on circuits, while providing a blueprint for analysing novel compositional inference queries., Comment: NeurIPS 2024

Adapting foundation models for specific purposes has become a standard approach to build machine learning systems for downstream applications. Yet, it is an open question which mechanisms take place during adaptation. Here we develop a new Sparse Autoencoder (SAE) for the CLIP vision transformer, named PatchSAE, to extract interpretable concepts at granular levels (e.g. shape, color, or semantics of an object) and their patch-wise spatial attributions. We explore how these concepts influence the model output in downstream image classification tasks and investigate how recent state-of-the-art prompt-based adaptation techniques change the association of model inputs to these concepts. While activations of concepts slightly change between adapted and non-adapted models, we find that the majority of gains on common adaptation tasks can be explained with the existing concepts already present in the non-adapted foundation model. This work provides a concrete framework to train and use SAEs for Vision Transformers and provides insights into explaining adaptation mechanisms., Comment: A demo is available at github.com/dynamical-inference/patchsae

This study presents the conditional neural fields for reduced-order modeling (CNF-ROM) framework to approximate solutions of parametrized partial differential equations (PDEs). The approach combines a parametric neural ODE (PNODE) for modeling latent dynamics over time with a decoder that reconstructs PDE solutions from the corresponding latent states. We introduce a physics-informed learning objective for CNF-ROM, which includes two key components. First, the framework uses coordinate-based neural networks to calculate and minimize PDE residuals by computing spatial derivatives via automatic differentiation and applying the chain rule for time derivatives. Second, exact initial and boundary conditions (IC/BC) are imposed using approximate distance functions (ADFs) [Sukumar and Srivastava, CMAME, 2022]. However, ADFs introduce a trade-off as their second- or higher-order derivatives become unstable at the joining points of boundaries. To address this, we introduce an auxiliary network inspired by [Gladstone et al., NeurIPS ML4PS workshop, 2022]. Our method is validated through parameter extrapolation and interpolation, temporal extrapolation, and comparisons with analytical solutions., Comment: 7 pages, 2 figures, NeurIPS 2024 Workshop on Machine Learning and the Physical Sciences

In the evolving landscape of machine learning (ML), Federated Learning (FL) presents a paradigm shift towards decentralized model training while preserving user data privacy. This paper introduces the concept of ``privacy drift", an innovative framework that parallels the well-known phenomenon of concept drift. While concept drift addresses the variability in model accuracy over time due to changes in the data, privacy drift encapsulates the variation in the leakage of private information as models undergo incremental training. By defining and examining privacy drift, this study aims to unveil the nuanced relationship between the evolution of model performance and the integrity of data privacy. Through rigorous experimentation, we investigate the dynamics of privacy drift in FL systems, focusing on how model updates and data distribution shifts influence the susceptibility of models to privacy attacks, such as membership inference attacks (MIA). Our results highlight a complex interplay between model accuracy and privacy safeguards, revealing that enhancements in model performance can lead to increased privacy risks. We provide empirical evidence from experiments on customized datasets derived from CIFAR-100 (Canadian Institute for Advanced Research, 100 classes), showcasing the impact of data and concept drift on privacy. This work lays the groundwork for future research on privacy-aware machine learning, aiming to achieve a delicate balance between model accuracy and data privacy in decentralized environments., Comment: 6 pages, 7 figures, Accepted in IEEE ICNC 25

We analyzed the 7.92$\times 10^{11}$ cosmic-ray-induced muon events collected by the IceCube Neutrino Observatory from May 13, 2011, when the fully constructed experiment started to take data, to May 12, 2023. This dataset provides an up-to-date cosmic-ray arrival direction distribution in the Southern Hemisphere with unprecedented statistical accuracy covering more than a full period length of a solar cycle. Improvements in Monte Carlo event simulation and better handling of year-to-year differences in data processing significantly reduce systematic uncertainties below the level of statistical fluctuations compared to the previously published results. We confirm the observation of a change in the angular structure of the cosmic-ray anisotropy between 10 TeV and 1 PeV, more specifically in the 100-300 TeV energy range.

In Laser Powder Bed Fusion (LPBF), the applied laser energy produces high thermal gradients that lead to unacceptable final part distortion. Accurate distortion prediction is essential for optimizing the 3D printing process and manufacturing a part that meets geometric accuracy requirements. This study introduces data-driven parameterized reduced-order models (ROMs) to predict distortion in LPBF across various machine process settings. We propose a ROM framework that combines Proper Orthogonal Decomposition (POD) with Gaussian Process Regression (GPR) and compare its performance against a deep-learning based parameterized graph convolutional autoencoder (GCA). The POD-GPR model demonstrates high accuracy, predicting distortions within $\pm0.001mm$, and delivers a computational speed-up of approximately 1800x., Comment: 7 pages, 4 figures, NeurIPS Machine Learning for Physical Sciences workshop