Search

Out-of-distribution (OOD) detection, which determines whether a given sample is part of the in-distribution (ID), has recently shown promising results through training with synthetic OOD datasets. Nonetheless, existing methods often produce outliers that are considerably distant from the ID, showing limited efficacy for capturing subtle distinctions between ID and OOD. To address these issues, we propose a novel framework, Semantic Outlier generation via Nuisance Awareness (SONA), which notably produces challenging outliers by directly leveraging pixel-space ID samples through diffusion models. Our approach incorporates SONA guidance, providing separate control over semantic and nuisance regions of ID samples. Thereby, the generated outliers achieve two crucial properties: (i) they present explicit semantic-discrepant information, while (ii) maintaining various levels of nuisance resemblance with ID. Furthermore, the improved OOD detector training with SONA outliers facilitates learning with a focus on semantic distinctions. Extensive experiments demonstrate the effectiveness of our framework, achieving an impressive AUROC of 88% on near-OOD datasets, which surpasses the performance of baseline methods by a significant margin of approximately 6%.

Multimodal Dialogue Response Generation (MDRG) is a recently proposed task where the model needs to generate responses in texts, images, or a blend of both based on the dialogue context. Due to the lack of a large-scale dataset specifically for this task and the benefits of leveraging powerful pre-trained models, previous work relies on the text modality as an intermediary step for both the image input and output of the model rather than adopting an end-to-end approach. However, this approach can overlook crucial information about the image, hindering 1) image-grounded text response and 2) consistency of objects in the image response. In this paper, we propose BI-MDRG that bridges the response generation path such that the image history information is utilized for enhanced relevance of text responses to the image content and the consistency of objects in sequential image responses. Through extensive experiments on the multimodal dialogue benchmark dataset, we show that BI-MDRG can effectively increase the quality of multimodal dialogue. Additionally, recognizing the gap in benchmark datasets for evaluating the image consistency in multimodal dialogue, we have created a curated set of 300 dialogues annotated to track object consistency across conversations., Comment: ECCV 2024

Test-Time Adaptation (TTA) has emerged as a crucial solution to the domain shift challenge, wherein the target environment diverges from the original training environment. A prime exemplification is TTA for Automatic Speech Recognition (ASR), which enhances model performance by leveraging output prediction entropy minimization as a self-supervision signal. However, a key limitation of this self-supervision lies in its primary focus on acoustic features, with minimal attention to the linguistic properties of the input. To address this gap, we propose Language Informed Test-Time Adaptation (LI-TTA), which incorporates linguistic insights during TTA for ASR. LI-TTA integrates corrections from an external language model to merge linguistic with acoustic information by minimizing the CTC loss from the correction alongside the standard TTA loss. With extensive experiments, we show that LI-TTA effectively improves the performance of TTA for ASR in various distribution shift situations., Comment: INTERSPEECH 2024

Reinforcement Learning from Human Feedback (RLHF) leverages human preference data to train language models to align more closely with human essence. These human preference data, however, are labeled at the sequence level, creating a mismatch between sequence-level preference labels and tokens, which are autoregressively generated from the language model. Although several recent approaches have tried to provide token-level (i.e., dense) rewards for each individual token, these typically rely on predefined discrete reward values (e.g., positive: +1, negative: -1, neutral: 0), failing to account for varying degrees of preference inherent to each token. To address this limitation, we introduce TLCR (Token-Level Continuous Reward) for RLHF, which incorporates a discriminator trained to distinguish positive and negative tokens, and the confidence of the discriminator is used to assign continuous rewards to each token considering the context. Extensive experiments show that our proposed TLCR leads to consistent performance improvements over previous sequence-level or token-level discrete rewards on open-ended generation benchmarks., Comment: ACL2024 Findings

Event cameras excel in capturing high-contrast scenes and dynamic objects, offering a significant advantage over traditional frame-based cameras. Despite active research into leveraging event cameras for semantic segmentation, generating pixel-wise dense semantic maps for such challenging scenarios remains labor-intensive. As a remedy, we present EV-WSSS: a novel weakly supervised approach for event-based semantic segmentation that utilizes sparse point annotations. To fully leverage the temporal characteristics of event data, the proposed framework performs asymmetric dual-student learning between 1) the original forward event data and 2) the longer reversed event data, which contain complementary information from the past and the future, respectively. Besides, to mitigate the challenges posed by sparse supervision, we propose feature-level contrastive learning based on class-wise prototypes, carefully aggregated at both spatial region and sample levels. Additionally, we further excavate the potential of our dual-student learning model by exchanging prototypes between the two learning paths, thereby harnessing their complementary strengths. With extensive experiments on various datasets, including DSEC Night-Point with sparse point annotations newly provided by this paper, the proposed method achieves substantial segmentation results even without relying on pixel-level dense ground truths. The code and dataset are available at https://github.com/Chohoonhee/EV-WSSS., Comment: Accepted by ECCV2024

We study the equilibria of uniform price auctions where bidders have flat demands up to their respective quantity constraints. We present an iterative procedure that systematically finds a Nash equilibrium outcome under semi-complete information as well as a novel ascending auction under incomplete information that has this outcome as a dominant strategy equilibrium. Demand reduction and low price equilibrium may occur since it is sometimes advantageous for a bidder to give up some of his/her demand and get the remaining demand at a low price rather than to get his/her entire demand at a higher price.

The $J/\psi$ and $\psi(2S)$ charmonium states, composed of $c\bar{c}$ quark pairs and known since the 1970s, are widely believed to serve as ideal probes to test quantum chromodynamics in high-energy hadronic interactions. However, there is not yet a complete understanding of the charmonium-production mechanism. Recent measurements of $J/\psi$ production as a function of event charged-particle multiplicity at the collision energies of both the Large Hadron Collider (LHC) and the Relativistic Heavy Ion Collider (RHIC) show enhanced $J/\psi$ production yields with increasing multiplicity. One potential explanation for this type of dependence is multiparton interactions (MPI). We carry out the first measurements of self-normalized $J/\psi$ yields and the $\psi(2S)$ to $J/\psi$ ratio at both forward and backward rapidities as a function of self-normalized charged-particle multiplicity in $p$$+$$p$ collisions at $\sqrt{s}=200$ GeV. In addition, detailed {\sc pythia} studies tuned to RHIC energies were performed to investigate the MPI impacts. We find that the PHENIX data at RHIC are consistent with recent LHC measurements and can only be described by {\sc pythia} calculations that include MPI effects. The forward and backward $\psi(2S)$ to $J/\psi$ ratio, which serves as a unique and powerful approach to study final-state effects on charmonium production, is found to be less dependent on the charged-particle multiplicity., Comment: 301 authors from 69 institutions, 8 pages, 3 figures. v1 is version submitted to Physical Review D Letters. HEPdata tables for the points plotted in figures for this and previous PHENIX publications are (or will be) publicly available at http://www.phenix.bnl.gov/papers.html

In situ observations by the Parker Solar Probe (PSP) have revealed new properties of the proton velocity distributions, including hammerhead features that suggest non-isotropic broadening of the beams. The present work proposes a very plausible explanation for the formation of these populations through the action of a proton firehose-like instability triggered by the proton beam. The quasi-linear (QL) theory proposed here shows that the resulting right-hand (RH) waves have two consequences on the protons: (i) reduce the relative drift between the beam and the core, but above all, (ii) induce a strong perpendicular temperature anisotropy, specific to the observed hammerhead ion strahl. Moreover, the long-run QL results suggest that these hammerhead distributions are rather transitory states, still subject to relaxation mechanisms, of which instabilities like the one discussed here are very likely involved.

After shipment to the Daresbury Lab and return to Fermilab, the prototype HB650 cryomodule underwent another phase of 2K RF testing to ascertain any performance issues that may have arisen from the transport of the cryomodule. While measurements taken at room temperature after the conclusion of shipment indicated that there were no negative impacts on cavity alignment, beamline vacuum, or cavity frequency, testing at 2K was required to validate other aspects such as tuner operation, cavity coupling, cryogenic system integrity, and cavity performance. Results of this latest round of limited 2K testing will be presented., Comment: 32nd Linear Accelerator Conference (LINAC 2024)

The rapid evolution of artificial intelligence, especially in large language models (LLMs), has significantly impacted various domains, including healthcare. In chest X-ray (CXR) analysis, previous studies have employed LLMs, but with limitations: either underutilizing the multi-tasking capabilities of LLMs or lacking clinical accuracy. This paper presents M4CXR, a multi-modal LLM designed to enhance CXR interpretation. The model is trained on a visual instruction-following dataset that integrates various task-specific datasets in a conversational format. As a result, the model supports multiple tasks such as medical report generation (MRG), visual grounding, and visual question answering (VQA). M4CXR achieves state-of-the-art clinical accuracy in MRG by employing a chain-of-thought prompting strategy, in which it identifies findings in CXR images and subsequently generates corresponding reports. The model is adaptable to various MRG scenarios depending on the available inputs, such as single-image, multi-image, and multi-study contexts. In addition to MRG, M4CXR performs visual grounding at a level comparable to specialized models and also demonstrates outstanding performance in VQA. Both quantitative and qualitative assessments reveal M4CXR's versatility in MRG, visual grounding, and VQA, while consistently maintaining clinical accuracy.

Video deblurring aims to enhance the quality of restored results in motion-blurred videos by effectively gathering information from adjacent video frames to compensate for the insufficient data in a single blurred frame. However, when faced with consecutively severe motion blur situations, frame-based video deblurring methods often fail to find accurate temporal correspondence among neighboring video frames, leading to diminished performance. To address this limitation, we aim to solve the video deblurring task by leveraging an event camera with micro-second temporal resolution. To fully exploit the dense temporal resolution of the event camera, we propose two modules: 1) Intra-frame feature enhancement operates within the exposure time of a single blurred frame, iteratively enhancing cross-modality features in a recurrent manner to better utilize the rich temporal information of events, 2) Inter-frame temporal feature alignment gathers valuable long-range temporal information to target frames, aggregating sharp features leveraging the advantages of the events. In addition, we present a novel dataset composed of real-world blurred RGB videos, corresponding sharp videos, and event data. This dataset serves as a valuable resource for evaluating event-guided deblurring methods. We demonstrate that our proposed methods outperform state-of-the-art frame-based and event-based motion deblurring methods through extensive experiments conducted on both synthetic and real-world deblurring datasets. The code and dataset are available at https://github.com/intelpro/CMTA., Comment: Accepted in ECCV2024

In low-light conditions, capturing videos with frame-based cameras often requires long exposure times, resulting in motion blur and reduced visibility. While frame-based motion deblurring and low-light enhancement have been studied, they still pose significant challenges. Event cameras have emerged as a promising solution for improving image quality in low-light environments and addressing motion blur. They provide two key advantages: capturing scene details well even in low light due to their high dynamic range, and effectively capturing motion information during long exposures due to their high temporal resolution. Despite efforts to tackle low-light enhancement and motion deblurring using event cameras separately, previous work has not addressed both simultaneously. To explore the joint task, we first establish real-world datasets for event-guided low-light enhancement and deblurring using a hybrid camera system based on beam splitters. Subsequently, we introduce an end-to-end framework to effectively handle these tasks. Our framework incorporates a module to efficiently leverage temporal information from events and frames. Furthermore, we propose a module to utilize cross-modal feature information to employ a low-pass filter for noise suppression while enhancing the main structural information. Our proposed method significantly outperforms existing approaches in addressing the joint task. Our project pages are available at https://github.com/intelpro/ELEDNet., Comment: Accepted in ECCV2024

We propose VoiceTailor, a parameter-efficient speaker-adaptive text-to-speech (TTS) system, by equipping a pre-trained diffusion-based TTS model with a personalized adapter. VoiceTailor identifies pivotal modules that benefit from the adapter based on a weight change ratio analysis. We utilize Low-Rank Adaptation (LoRA) as a parameter-efficient adaptation method and incorporate the adapter into pivotal modules of the pre-trained diffusion decoder. To achieve powerful adaptation performance with few parameters, we explore various guidance techniques for speaker adaptation and investigate the best strategies to strengthen speaker information. VoiceTailor demonstrates comparable speaker adaptation performance to existing adaptive TTS models by fine-tuning only 0.25\% of the total parameters. VoiceTailor shows strong robustness when adapting to a wide range of real-world speakers, as shown in the demo., Comment: INTERSPEECH 2024

Activation sparsity can enable practical inference speedups in large language models (LLMs) by reducing the compute and memory-movement required for matrix multiplications during the forward pass. However, existing methods face limitations that inhibit widespread adoption. Some approaches are tailored towards older models with ReLU-based sparsity, while others require extensive continued pre-training on up to hundreds of billions of tokens. This paper describes TEAL, a simple training-free method that applies magnitude-based activation sparsity to hidden states throughout the entire model. TEAL achieves 40-50% model-wide sparsity with minimal performance degradation across Llama-2, Llama-3, and Mistral families, with sizes varying from 7B to 70B. We improve existing sparse kernels and demonstrate wall-clock decoding speed-ups of up to 1.53$\times$ and 1.8$\times$ at 40% and 50% model-wide sparsity. TEAL is compatible with weight quantization, enabling further efficiency gains.

Existing portrait relighting methods struggle with precise control over facial shadows, particularly when faced with challenges such as handling hard shadows from directional light sources or adjusting shadows while remaining in harmony with existing lighting conditions. In many situations, completely altering input lighting is undesirable for portrait retouching applications: one may want to preserve some authenticity in the captured environment. Existing shadow editing methods typically restrict their application to just the facial region and often offer limited lighting control options, such as shadow softening or rotation. In this paper, we introduce COMPOSE: a novel shadow editing pipeline for human portraits, offering precise control over shadow attributes such as shape, intensity, and position, all while preserving the original environmental illumination of the portrait. This level of disentanglement and controllability is obtained thanks to a novel decomposition of the environment map representation into ambient light and an editable gaussian dominant light source. COMPOSE is a four-stage pipeline that consists of light estimation and editing, light diffusion, shadow synthesis, and finally shadow editing. We define facial shadows as the result of a dominant light source, encoded using our novel gaussian environment map representation. Utilizing an OLAT dataset, we have trained models to: (1) predict this light source representation from images, and (2) generate realistic shadows using this representation. We also demonstrate comprehensive and intuitive shadow editing with our pipeline. Through extensive quantitative and qualitative evaluations, we have demonstrated the robust capability of our system in shadow editing., Comment: Accepted at ECCV 2024

This study presents a cross-verification of fast ion effects on turbulence through a systematic comparison of two leading gyrokinetic codes, GENE [T.Gorler et al., J. Comput. Phys. 230 7053-7071 (2011)] and CGYRO [J.Candy et al, J. Comput. Phys. 324 73-93 (2016)], using L-mode plasma profiles from KSTAR for local linear and nonlinear electromagnetic simulations. The focus is on the impact of fast ions and rotation effects on energy flux, aiming to identify the similarities and differences between these codes in the context of turbulence transport research. The analysis shows consistency in linear stability results, fractional changes in energy flux, and zonal shearing between the codes. However, discrepancies arise in absolute thermal energy levels, phase angle distribution, and rotation effects on energy transport, especially in the presence of fast ions. The study underscores the critical importance of phase angle analysis in gyrokinetic code verification, particularly when assessing fast ion effects on turbulence. Additionally, it highlights the need to examine quantities at lower levels of the primacy hierarchy, as discrepancies at higher levels can lead to divergent results at lower levels. These findings indicate the necessity for further investigation into these discrepancies and the novel phase angle structures observed, contributing to the advancement of accurate transport predictions in fusion plasmas.

Uplift modeling is essential for optimizing marketing strategies by selecting individuals likely to respond positively to specific marketing campaigns. This importance escalates in multi-treatment marketing campaigns, where diverse treatment is available and we may want to assign the customers to treatment that can make the most impact. While there are existing approaches with convenient frameworks like Causalml, there are potential spaces to enhance the effect of uplift modeling in multi treatment cases. This paper introduces a novel approach to uplift modeling in multi-treatment campaigns, leveraging score ranking and calibration techniques to improve overall performance of the marketing campaign. We review existing uplift models, including Meta Learner frameworks (S, T, X), and their application in real-world scenarios. Additionally, we delve into insights from multi-treatment studies to highlight the complexities and potential advancements in the field. Our methodology incorporates Meta-Learner calibration and a scoring rank-based offer selection strategy. Extensive experiment results with real-world datasets demonstrate the practical benefits and superior performance of our approach. The findings underscore the critical role of integrating score ranking and calibration techniques in refining the performance and reliability of uplift predictions, thereby advancing predictive modeling in marketing analytics and providing actionable insights for practitioners seeking to optimize their campaign strategies.

Few-shot Anomaly Detection (FAD) poses significant challenges due to the limited availability of training samples and the frequent absence of abnormal samples. Previous approaches often rely on annotations or true abnormal samples to improve detection, but such textual or visual cues are not always accessible. To address this, we introduce AnoPLe, a multi-modal prompt learning method designed for anomaly detection without prior knowledge of anomalies. AnoPLe simulates anomalies and employs bidirectional coupling of textual and visual prompts to facilitate deep interaction between the two modalities. Additionally, we integrate a lightweight decoder with a learnable multi-view signal, trained on multi-scale images to enhance local semantic comprehension. To further improve performance, we align global and local semantics, enriching the image-level understanding of anomalies. The experimental results demonstrate that AnoPLe achieves strong FAD performance, recording 94.1% and 86.2% Image AUROC on MVTec-AD and VisA respectively, with only around a 1% gap compared to the SoTA, despite not being exposed to true anomalies. Code is available at https://github.com/YoojLee/AnoPLe., Comment: Code is available at https://github.com/YoojLee/AnoPLe

We introduce a simple parametric model of the radio-infrared correlation (i.e., the ratio between the IR luminosity and the 1.4 GHz radio luminosity, $q_{\mbox{\tiny IR}}$) by considering the energy loss rate of high-energy cosmic ray (CR) electron governed by the radiative cooling (synchrotron, bremsstrahlung, inverse Compton scattering), ionization, and adiabatic expansion. Each process of CR electron energy loss is explicitly computed and compared to each other. We rewrite the energy loss rate of each process to be dependent on the gas surface density and redshift using the relevant scaling relations. By combining each energy loss rate, the fraction of the synchrotron energy loss rate is computed as a function of gas surface density and redshift, and used to extrapolate the well-established `local' radio-infrared correlation to the high-redshift universe. The locally established $q_{\mbox{\tiny IR}}$ is reformulated to be dependent upon the redshift and the gas surface density and applied for understanding the observed distribution of the radio-infrared correlation of high-redshift galaxies in \cite{delvecchio_etal_2021}. Our model predicts that $q_{\mbox{\tiny IR}}$ value is anti-correlated with gas surface density and the redshift dependency of $q_{\mbox{\tiny IR}}$ value changes by gas surface density of galaxies, which captures the observed trend of $q_{\mbox{\tiny IR}}$ values for stellar mass selected star forming galaxies with a minimal impact of radio-infrared selection bias., Comment: 13 pages, 6 figures, accepted for publication in ApJ

Instead of randomly acquiring training data points, Uncertainty-based Active Learning (UAL) operates by querying the label(s) of pivotal samples from an unlabeled pool selected based on the prediction uncertainty, thereby aiming at minimizing the labeling cost for model training. The efficacy of UAL critically depends on the model capacity as well as the adopted uncertainty-based acquisition function. Within the context of this study, our analytical focus is directed toward comprehending how the capacity of the machine learning model may affect UAL efficacy. Through theoretical analysis, comprehensive simulations, and empirical studies, we conclusively demonstrate that UAL can lead to worse performance in comparison with random sampling when the machine learning model class has low capacity and is unable to cover the underlying ground truth. In such situations, adopting acquisition functions that directly target estimating the prediction performance may be beneficial for improving the performance of UAL.

The intent of this paper is to explore Dowker duality from a combinatorial, topological, and categorical perspective. The paper presents three short, new proofs of Dowker duality using various poset fiber lemmas. We introduce modifications of joins and products of simplicial complexes called relational join and relational product complexes. These relational complexes can be constructed whenever there is a relation between simplicial complexes, which includes the context of Dowker duality and covers of simplicial complexes. In this more general setting, we show that the homologies of the simplicial complexes and the relational complexes fit together in a long exact sequence. Similar results are then established for profunctors, which are generalizations of relations to categories. The cograph and graph of profunctors play the role of the relational join and relational product complexes. For a profunctor that arise from adjoint functors between $C$ and $D$, we show that $C$, $D$, the cograph, and the graph all have homotopy equivalent classifying spaces. Lastly, we show that given any profunctor from $D$ to $C$, the homologies of $C$, $D$, the cograph, and the graph form a long exact sequence. We make use of a spectral sequence for Grothendieck fibrations of small categories.

Unlike fixed- or variable-rate image coding, progressive image coding (PIC) aims to compress various qualities of images into a single bitstream, increasing the versatility of bitstream utilization and providing high compression efficiency compared to simulcast compression. Research on neural network (NN)-based PIC is in its early stages, mainly focusing on applying varying quantization step sizes to the transformed latent representations in a hierarchical manner. These approaches are designed to compress only the progressively added information as the quality improves, considering that a wider quantization interval for lower-quality compression includes multiple narrower sub-intervals for higher-quality compression. However, the existing methods are based on handcrafted quantization hierarchies, resulting in sub-optimal compression efficiency. In this paper, we propose an NN-based progressive coding method that firstly utilizes learned quantization step sizes via learning for each quantization layer. We also incorporate selective compression with which only the essential representation components are compressed for each quantization layer. We demonstrate that our method achieves significantly higher coding efficiency than the existing approaches with decreased decoding time and reduced model size.

Recent text-to-image models like Stable Diffusion produce photo-realistic images but often show demographic biases. Previous debiasing methods focused on training-based approaches, failing to explore the root causes of bias and overlooking Stable Diffusion's potential for unbiased image generation. In this paper, we demonstrate that Stable Diffusion inherently possesses fairness, which can be unlocked to achieve debiased outputs. Through carefully designed experiments, we identify the excessive bonding between text prompts and the diffusion process as a key source of bias. To address this, we propose a novel approach that perturbs text conditions to unleash Stable Diffusion's intrinsic fairness. Our method effectively mitigates bias without additional tuning, while preserving image-text alignment and image quality., Comment: 21 pages, 20 figures; First two authors contributed equally

Building on previous works to construct a homogeneous sample of brown dwarfs in binary systems, we investigate microlensing events detected by the Korea Microlensing Telescope Network (KMTNet) survey during the 2022 and 2023 seasons. Given the difficulty in distinguishing brown-dwarf events from those produced by binary lenses with nearly equal-mass components, we analyze all lensing events detected during the seasons that exhibit anomalies characteristic of binary-lens systems. Using the same criteria consistently applied in previous studies, we identify six additional brown dwarf candidates through the analysis of lensing events KMT-2022-BLG-0412, KMT-2022-BLG-2286, KMT-2023-BLG-0201, KMT-2023-BLG-0601, KMT-2023-BLG-1684, and KMT-2023-BLG-1743. An examination of the mass posteriors shows that the median mass of the lens companions ranges from 0.02 $M_\odot$ to 0.05 $M_\odot$, indicating that these companions fall within the brown-dwarf mass range. The mass of the primary lenses ranges from 0.11 $M_\odot$ to 0.68 $M_\odot$, indicating that they are low-mass stars with substantially lower masses compared to the Sun., Comment: 13 pages, 17 figures, 12 tables

Topology profoundly influences diverse fields of science, providing a powerful framework for classifying phases of matter and predicting nontrivial excitations, such as solitons, vortices, and skyrmions. These topological defects are typically characterized by integer numbers, called topological charges, representing the winding number in their order parameter field. The classification and prediction of topological defects, however, become challenging when singularities are included within the integration domain for calculating the topological charge. While such exotic nonlinear excitations have been proposed in the superfluid $^3$He-A phase and spinor Bose-Einstein condensate of atomic gases, experimental observation of these structures and studies of their stability have long been elusive. Here we report the observation of a singular skyrmion that goes beyond the framework of topology in a ferromagnetic superfluid. The exotic skyrmions are sustained by undergoing anomalous symmetry breaking associated with the eccentric spin singularity and carry half of the elementary charge, distinctive from conventional skyrmions or merons. By successfully realizing the universal regime of the quantum Kelvin-Helmholtz instability, we identified the eccentric fractional skyrmions, produced by emission from a magnetic domain wall and a spontaneous splitting of an integer skyrmion with spin singularities. The singular skyrmions are stable and can be observed after 2~s of hold time. Our results confirm the universality between classical and quantum Kelvin-Helmholtz instabilities and broaden our understanding on complex nonlinear dynamics of nontrivial texture beyond skyrmion in topological quantum systems., Comment: 13 pages, 5 main figures and 7 supplemental figures

The jet cross-section and jet-substructure observables in $p$$+$$p$ collisions at $\sqrt{s}=200$ GeV were measured by the PHENIX Collaboration at the Relativistic Heavy Ion Collider (RHIC). Jets are reconstructed from charged-particle tracks and electromagnetic-calorimeter clusters using the anti-$k_{t}$ algorithm with a jet radius $R=0.3$ for jets with transverse momentum within $8.0

High-resolution Vision-Language Models (VLMs) have been widely used in multimodal tasks to enhance accuracy by preserving detailed image information. However, these models often generate excessive visual tokens due to encoding multiple partitions of the input image. Processing these excessive visual tokens is computationally challenging, especially in resource-constrained environments with commodity GPUs. To support high-resolution images while meeting resource constraints, we propose High-Resolution Early Dropping (HiRED), a token-dropping scheme that operates within a fixed token budget before the Large Language Model (LLM) stage. HiRED can be integrated with existing high-resolution VLMs in a plug-and-play manner, as it requires no additional training while still maintaining superior accuracy. We strategically use the vision encoder's attention in the initial layers to assess the visual content of each image partition and allocate the token budget accordingly. Then, using the attention in the final layer, we select the most important visual tokens from each partition within the allocated budget, dropping the rest. Empirically, when applied to LLaVA-Next-7B on NVIDIA TESLA P40 GPU, HiRED with a 20% token budget increases token generation throughput by 4.7, reduces first-token generation latency by 15 seconds, and saves 2.3 GB of GPU memory for a single inference., Comment: Preprint

Many spatiotemporal domains handle multi-agent trajectory data, but in real-world scenarios, collected trajectory data are often partially missing due to various reasons. While existing approaches demonstrate good performance in trajectory imputation, they face challenges in capturing the complex dynamics and interactions between agents due to a lack of physical constraints that govern realistic trajectories, leading to suboptimal results. To address this issue, the paper proposes a Derivative-Based Hybrid Prediction (DBHP) framework that can effectively impute multiple agents' missing trajectories. First, a neural network equipped with Set Transformers produces a naive prediction of missing trajectories while satisfying the permutation-equivariance in terms of the order of input agents. Then, the framework makes alternative predictions leveraging velocity and acceleration information and combines all the predictions with properly determined weights to provide final imputed trajectories. In this way, our proposed framework not only accurately predicts position, velocity, and acceleration values but also enforces the physical relationship between them, eventually improving both the accuracy and naturalness of the predicted trajectories. Accordingly, the experiment results about imputing player trajectories in team sports show that our framework significantly outperforms existing imputation baselines.

Engineering non-Abelian quasiparticles by combining superconductivity and topological states have been proposed as a route to realize topological quantum computation. Rhombohedral multilayer graphene with layer number N>=3 has been shown as a promising platform, as it hosts integer and fractional quantum anomalous Hall effects when proximitized by transition metal dichalcogenide (TMD) and a moire potential. However, superconductivity in similar devices have remained largely unexplored, although proximitized spin-orbit-coupling (SOC) effect has been shown to strengthen or induce superconductivity in both crystalline and twisted graphene. Here we report electron transport measurements of TMD-proximitized rhombohedral trilayer graphene (RTG) at temperatures down to 40 mK. We observed a new hole-doped superconducting state SC4 with a transition temperature Tc of 230 mK. On the electron-doped side, we identified a new isospin-symmetry breaking three-quarter-metal (TQM) phase. Near this three-quarter-metal state, the state SC3, very weak in bare RTG, is fully developed into a superconducting state at 110 mK. By performing fermiology analysis based on the quantum oscillation measurement, we showed that the SC3 and SC4 states reside at the phase boundaries between different isospin-symmetry-breaking states. These observations are aligned with the existing understanding that SOC enhances graphene superconductivity. Surprisingly, the original superconducting state SC1 in bare RTG is strongly suppressed in the presence of TMD, and we cannot find it down to the base temperature of our measurement. Our observations form the basis of exploring superconductivity and non-Abelian quasiparticles in rhombohedral graphene devices, and provide experimental evidence that challenges the understanding of the impacts of SOC on graphene superconductivity., Comment: 35 pages; 4 figures, 1 table, 13 extended data figures

Artificial intelligence applied to retinal images offers significant potential for recognizing signs and symptoms of retinal conditions and expediting the diagnosis of eye diseases and systemic disorders. However, developing generalized artificial intelligence models for medical data often requires a large number of labeled images representing various disease signs, and most models are typically task-specific, focusing on major retinal diseases. In this study, we developed a Fundus-Specific Pretrained Model (Image+Fundus), a supervised artificial intelligence model trained to detect abnormalities in fundus images. A total of 57,803 images were used to develop this pretrained model, which achieved superior performance across various downstream tasks, indicating that our proposed model outperforms other general methods. Our Image+Fundus model offers a generalized approach to improve model performance while reducing the number of labeled datasets required. Additionally, it provides more disease-specific insights into fundus images, with visualizations generated by our model. These disease-specific foundation models are invaluable in enhancing the performance and efficiency of deep learning models in the field of fundus imaging., Comment: 10 pages, 4 figures

Language Models (LMs) are increasingly employed in recommendation systems due to their advanced language understanding and generation capabilities. Recent recommender systems based on generative retrieval have leveraged the inferential abilities of LMs to directly generate the index tokens of the next item, based on item sequences within the user's interaction history. Previous studies have mostly focused on item indices based solely on textual semantic or collaborative information. However, although the standalone effectiveness of these aspects has been demonstrated, the integration of this information has remained unexplored. Our in-depth analysis finds that there is a significant difference in the knowledge captured by the model from heterogeneous item indices and diverse input prompts, which can have a high potential for complementarity. In this paper, we propose SC-Rec, a unified recommender system that learns diverse preference knowledge from two distinct item indices and multiple prompt templates. Furthermore, SC-Rec adopts a novel reranking strategy that aggregates a set of ranking results, inferred based on different indices and prompts, to achieve the self-consistency of the model. Our empirical evaluation on three real-world datasets demonstrates that SC-Rec considerably outperforms the state-of-the-art methods for sequential recommendation, effectively incorporating complementary knowledge from varied outputs of the model.

The First Large Absorption Survey in HI (FLASH) is a large-area radio survey for neutral hydrogen in the redshift range 0.41$, and appear to be a mixture of intervening and associated systems. The overall detection rate for HI absorption lines in the Pilot Surveys (0.3 to 0.5 lines per ASKAP field) is a factor of two below the expected value. There are several possible reasons for this, but one likely factor is the presence of a range of spectral-line artefacts in the Pilot Survey data that have now been mitigated and are not expected to recur in the full FLASH survey. A future paper will discuss the host galaxies of the HI absorption systems identified here., Comment: 46 pages, 25 figures, 10 tables. Submitted to PASA

Diffusion models have shown remarkable performance in generation problems over various domains including images, videos, text, and audio. A practical bottleneck of diffusion models is their sampling speed, due to the repeated evaluation of score estimation networks during the inference. In this work, we propose a novel framework capable of adaptively allocating compute required for the score estimation, thereby reducing the overall sampling time of diffusion models. We observe that the amount of computation required for the score estimation may vary along the time step for which the score is estimated. Based on this observation, we propose an early-exiting scheme, where we skip the subset of parameters in the score estimation network during the inference, based on a time-dependent exit schedule. Using the diffusion models for image synthesis, we show that our method could significantly improve the sampling throughput of the diffusion models without compromising image quality. Furthermore, we also demonstrate that our method seamlessly integrates with various types of solvers for faster sampling, capitalizing on their compatibility to enhance overall efficiency. The source code and our experiments are available at \url{https://github.com/taehong-moon/ee-diffusion}, Comment: ICML 2024

Neurons integrate synaptic inputs and convert them to action potential output at electrically distant locations. The computational power of a neuron is hence enhanced by subcellular compartmentalization and nonlinear synaptic integration, but the biophysical determinants of these features in human neurons are not completely understood. By examining the synaptic input-output function of human neocortical pyramidal neurons, we found that the nonlinearity threshold at the soma was linearly determined by the shortest path distance from the synapse to the apical trunk, and the slope of this relationship was consistent throughout the dendritic arbor. Analogous rules were found from both supragranular and infragranular layers of the rodent cortex, suggesting that these represent a fundamental property of pyramidal neurons. Additionally, we found that neurons associated with tumor or epilepsy had distinct membrane properties, but the nonlinearity threshold was shifted in amplitude such that the slope of its relationship with synaptic distance remained consistent.

Accurate patient selection is critical in radiotherapy (RT) to prevent ineffective treatments. Traditional survival prediction models, relying on structured data, often lack precision. This study explores the potential of large language models (LLMs) to structure unstructured electronic health record (EHR) data, thereby improving survival prediction accuracy through comprehensive clinical information integration. Data from 34,276 patients treated with RT at Yonsei Cancer Center between 2013 and 2023 were analyzed, encompassing both structured and unstructured data. An open-source LLM was used to structure the unstructured EHR data via single-shot learning, with its performance compared against a domain-specific medical LLM and a smaller variant. Survival prediction models were developed using statistical, machine learning, and deep learning approaches, incorporating both structured and LLM-structured data. Clinical experts evaluated the accuracy of the LLM-structured data. The open-source LLM achieved 87.5% accuracy in structuring unstructured EHR data without additional training, significantly outperforming the domain-specific medical LLM, which reached only 35.8% accuracy. Larger LLMs were more effective, particularly in extracting clinically relevant features like general condition and disease extent, which closely correlated with patient survival. Incorporating LLM-structured clinical features into survival prediction models significantly improved accuracy, with the C-index of deep learning models increasing from 0.737 to 0.820. These models also became more interpretable by emphasizing clinically significant factors. This study shows that general-domain LLMs, even without specific medical training, can effectively structure large-scale unstructured EHR data, substantially enhancing the accuracy and interpretability of clinical predictive models., Comment: 23 pages, 2 tables, 4 figures

Retrieval-Augmented Language Models (RALMs) have significantly improved performance in open-domain question answering (QA) by leveraging external knowledge. However, RALMs still struggle with unanswerable queries, where the retrieved contexts do not contain the correct answer, and with conflicting information, where different sources provide contradictory answers due to imperfect retrieval. This study introduces an in-context learning-based approach to enhance the reasoning capabilities of RALMs, making them more robust in imperfect retrieval scenarios. Our method incorporates Machine Reading Comprehension (MRC) demonstrations, referred to as cases, to boost the model's capabilities to identify unanswerabilities and conflicts among the retrieved contexts. Experiments on two open-domain QA datasets show that our approach increases accuracy in identifying unanswerable and conflicting scenarios without requiring additional fine-tuning. This work demonstrates that in-context learning can effectively enhance the robustness of RALMs in open-domain QA tasks., Comment: 10 pages, 2 figures

The utilization of automated depression detection significantly enhances early intervention for individuals experiencing depression. Despite numerous proposals on automated depression detection using recorded clinical interview videos, limited attention has been paid to considering the hierarchical structure of the interview questions. In clinical interviews for diagnosing depression, clinicians use a structured questionnaire that includes routine baseline questions and follow-up questions to assess the interviewee's condition. This paper introduces HiQuE (Hierarchical Question Embedding network), a novel depression detection framework that leverages the hierarchical relationship between primary and follow-up questions in clinical interviews. HiQuE can effectively capture the importance of each question in diagnosing depression by learning mutual information across multiple modalities. We conduct extensive experiments on the widely-used clinical interview data, DAIC-WOZ, where our model outperforms other state-of-the-art multimodal depression detection models and emotion recognition models, showcasing its clinical utility in depression detection., Comment: 11 pages, 6 figures, Proceedings of the 33rd ACM International Conference on Information and Knowledge Management (CIKM '24)

Customers reach out to online live chat agents with various intents, such as asking about product details or requesting a return. In this paper, we propose the problem of predicting user intent from browsing history and address it through a two-stage approach. The first stage classifies a user's browsing history into high-level intent categories. Here, we represent each browsing history as a text sequence of page attributes and use the ground-truth class labels to fine-tune pretrained Transformers. The second stage provides a large language model (LLM) with the browsing history and predicted intent class to generate fine-grained intents. For automatic evaluation, we use a separate LLM to judge the similarity between generated and ground-truth intents, which closely aligns with human judgments. Our two-stage approach yields significant performance gains compared to generating intents without the classification stage., Comment: CIKM 2024

Identifying patient subgroups with different treatment responses is an important task to inform medical recommendations, guidelines, and the design of future clinical trials. Existing approaches for subgroup analysis primarily focus on Randomised Controlled Trials (RCTs), in which treatment assignment is randomised. Furthermore, the patient cohort of an RCT is often constrained by cost, and is not representative of the heterogeneity of patients likely to receive treatment in real-world clinical practice. Therefore, when applied to observational studies, such approaches suffer from significant statistical biases because of the non-randomisation of treatment. Our work introduces a novel, outcome-guided method for identifying treatment response subgroups in observational studies. Our approach assigns each patient to a subgroup associated with two time-to-event distributions: one under treatment and one under control regime. It hence positions itself in between individualised and average treatment effect estimation. The assumptions of our model result in a simple correction of the statistical bias from treatment non-randomisation through inverse propensity weighting. In experiments, our approach significantly outperforms the current state-of-the-art method for outcome-guided subgroup analysis in both randomised and observational treatment regimes.

In this study, we evaluate the predictive power of density functional theory (DFT) for the magnetic properties of MnBi\(_2\)Te\(_4\) (MBT), an intrinsically magnetic topological insulator with potential applications in spintronics and quantum computing. Our theoretical understanding of MBT has been challenged by discrepancies between experimental results and \textit{ab initio} calculations, particularly with respect to its electronic and magnetic properties. Our results show that the magnetic phase diagram of MBT varies significantly depending on the Hubbard $U$ parameter in the DFT framework, highlighting the importance of benchmark calculations. To address these challenges, we establish an optimized Hubbard $U$ approach derived from Diffusion Monte Carlo (DMC) calculations, which directly solves the many-body Schr\"{o}dinger equation based on the stochastic process, and implement it in the DFT framework. Once the optimized $U$ value is determined as a function of strain, we apply it to achieve DMC-level accuracy within our DFT framework. This approach is instrumental in accurately describing the magnetic states of MBT and understanding the underlying mechanisms governing its magnetic properties and their dependence on external factors., Comment: 8 pages, 5 figures

In this paper, we address the problem of unsupervised video anomaly detection (UVAD). The task aims to detect abnormal events in test video using unlabeled videos as training data. The presence of anomalies in the training data poses a significant challenge in this task, particularly because they form clusters in the feature space. We refer to this property as the "Anomaly Cluster" issue. The condensed nature of these anomalies makes it difficult to distinguish between normal and abnormal data in the training set. Consequently, training conventional anomaly detection techniques using an unlabeled dataset often leads to sub-optimal results. To tackle this difficulty, we propose a new method called Cleansed k-Nearest Neighbor (CKNN), which explicitly filters out the Anomaly Clusters by cleansing the training dataset. Following the k-nearest neighbor algorithm in the feature space provides powerful anomaly detection capability. Although the identified Anomaly Cluster issue presents a significant challenge to applying k-nearest neighbor in UVAD, our proposed cleansing scheme effectively addresses this problem. We evaluate the proposed method on various benchmark datasets and demonstrate that CKNN outperforms the previous state-of-the-art UVAD method by up to 8.5% (from 82.0 to 89.0) in terms of AUROC. Moreover, we emphasize that the performance of the proposed method is comparable to that of the state-of-the-art method trained using anomaly-free data.

In the present paper, firstly we obtain the canal hypersurfaces that are formed as the envelope of a family of pseudo hyperspheres or pseudo hyperbolic hyperspheres whose centers lie on a spacelike curve with parallel timelike normal vector field $B_{2}$ in Minkowski space-time and we give some geometric characterizations for them by obtaining the Gaussian curvature, mean curvature and principal curvatures of these canal hypersurfaces. Also, we give the general expression of parametrizations of the canal hypersurfaces generated by non-null curves with parallel frame in $E_{1}^{4}$ and we obtain some important geometric invariants and characterizations for them.

In this work, we investigate the electrical conductivity of carbon nanotubes (CNTs), with a particular focus on the effects of doping. Using a first-principles approach, we study the electronic structure, phonon dispersion, and electron-phonon scattering to understand the finite-temperature electrical transport properties in CNTs. Our study covers both prototypical metallic and semiconducting CNTs, with special emphasis on the influence of typical defects such as vacancies and the incorporation of copper or nitrogen, such as pyridinic N, pyrrolic N, graphitic N, and oxidized N. Our theoretical study shows significant improvements in the electrical conduction properties of copper-CNT composites, especially when semiconducting CNTs are functionalized with nitrogen. Doping is found to cause significant changes in the electronic density of states near the Fermi level, which affects the electrical conductivity. Calculations show that certain types of functional groups, such as N-pyrrolic, result in a ~30-fold increase in the conductivity of semiconducting CNTs compared to Cu-incorporated CNTs alone. For metallic CNTs, the conductivity is in agreement with existing experimental data, and our prediction of significant increases in conductivity with N-pyrrolic functional group is consistent with recent experimental results, demonstrating the effectiveness of doping in modifying conductivity. Our study provides valuable insight into the electronic properties of doped CNTs and contributes to the development of ultra-high conductivity CNT composites., Comment: 16 pages, 4 figures, 39 references

Humans are not homo economicus (i.e., rational economic beings). As humans, we exhibit systematic behavioral biases such as loss aversion, anchoring, framing, etc., which lead us to make suboptimal economic decisions. Insofar as such biases may be embedded in text data on which large language models (LLMs) are trained, to what extent are LLMs prone to the same behavioral biases? Understanding these biases in LLMs is crucial for deploying LLMs to support human decision-making. We propose utility theory-a paradigm at the core of modern economic theory-as an approach to evaluate the economic biases of LLMs. Utility theory enables the quantification and comparison of economic behavior against benchmarks such as perfect rationality or human behavior. To demonstrate our approach, we quantify and compare the economic behavior of a variety of open- and closed-source LLMs. We find that the economic behavior of current LLMs is neither entirely human-like nor entirely economicus-like. We also find that most current LLMs struggle to maintain consistent economic behavior across settings. Finally, we illustrate how our approach can measure the effect of interventions such as prompting on economic biases., Comment: Accepted to COLM 2024

Calibrated recommendation, which aims to maintain personalized proportions of categories within recommendations, is crucial in practical scenarios since it enhances user satisfaction by reflecting diverse interests. However, achieving calibration in a sequential setting (i.e., calibrated sequential recommendation) is challenging due to the need to adapt to users' evolving preferences. Previous methods typically leverage reranking algorithms to calibrate recommendations after training a model without considering the effect of calibration and do not effectively tackle the conflict between relevance and calibration during the reranking process. In this work, we propose LeapRec (Calibration-Disentangled Learning and Relevance-Prioritized Reranking), a novel approach for the calibrated sequential recommendation that addresses these challenges. LeapRec consists of two phases, model training phase and reranking phase. In the training phase, a backbone model is trained using our proposed calibration-disentangled learning-to-rank loss, which optimizes personalized rankings while integrating calibration considerations. In the reranking phase, relevant items are prioritized at the top of the list, with items needed for calibration following later to address potential conflicts between relevance and calibration. Through extensive experiments on four real-world datasets, we show that LeapRec consistently outperforms previous methods in the calibrated sequential recommendation. Our code is available at https://github.com/jeon185/LeapRec., Comment: Published at CIKM '24 as a full research paper