arXiv subjects

Gang Wu

Publications and source records attributed to Gang Wu.

At least 73 records · Page 4Linked to original sources

arXiv2025

TickIt: Leveraging Large Language Models for Automated Ticket Escalation

In large-scale cloud service systems, support tickets serve as a critical mechanism for resolving customer issues and maintaining service quality. However, traditional manual ticket escalation processes encounter significant challenges, including inefficiency, inaccuracy, and difficulty in handling the high volume and complexity of tickets. While previous research has proposed various machine learning models for ticket classification, these approaches often overlook the practical demands of real-world escalations, such as dynamic ticket updates, topic-specific routing, and the analysis of ticket relationships. To bridge this gap, this paper introduces TickIt, an innovative online ticket escalation framework powered by Large Language Models. TickIt enables topic-aware, dynamic, and relationship-driven ticket escalations by continuously updating ticket states, assigning tickets to the most appropriate support teams, exploring ticket correlations, and leveraging category-guided supervised fine-tuning to continuously improve its performance. By deploying TickIt in ByteDance's cloud service platform Volcano Engine, we validate its efficacy and practicality, marking a significant advancement in the field of automated ticket escalation for large-scale cloud service systems.

cs.SE↗

arXiv2025

Stellar tidal streams around nearby spiral galaxies with deep imaging from amateur telescopes

Tidal interactions between massive galaxies and their satellites are fundamental processes in a Universe with L-Cold Dark Matter cosmology, redistributing material into faint features that preserve records of past galactic interactions. While stellar streams in the Local Group impressively demonstrate satellite disruption, they do not constitute a statistically significant sample. Constructing a substantial catalog of stellar streams beyond the Local Group remains challenging due to the difficulties in obtaining deep, wide-field images of galaxies. Despite their potential to illuminate dark matter distribution and galaxy formation processes, stellar streams remain underutilized as cosmological probes. The Stellar Tidal Stream Survey (STSS) addresses this observational gap by leveraging amateur telescopes to obtain deep, scientific-grade images of galactic outskirts, capable of building a more statistically meaningful sample of stellar streams. Over the last decade, the STSS has acquired deep (up to surface brightness limit 28.3 mag/arcsec^2 in the r-band), wide-field images of 15 nearby Milky Way analog galaxies using a network of robotic amateur telescopes, avoiding the issues associated with "mosaicing" smaller images taken with a professional telescope. Our survey has revealed a diverse range of previously unreported faint features related to dwarf satellite accretion - including stellar streams, shells, and umbrella-like structures. We discover an ultra-diffuse galaxy (NGC150-UDG1), which shows hints of tidal tails. The STSS demonstrates the suitability of modern amateur telescopes to detect and study faint, diffuse structures in large fields around nearby spiral galaxies. Their economic and accessibility advantages enable larger samples with deep imaging, essential for testing galaxy formation models and constraining the properties of minor merger events in the local Universe.

astro-ph.GA↗

arXiv2025

ScreenLLM: Stateful Screen Schema for Efficient Action Understanding and Prediction

Graphical User Interface (GUI) agents are autonomous systems that interpret and generate actions, enabling intelligent user assistance and automation. Effective training of these agent presents unique challenges, such as sparsity in supervision signals, scalability for large datasets, and the need for nuanced user understanding. We propose stateful screen schema, an efficient representation of GUI interactions that captures key user actions and intentions over time. Building on this foundation, we introduce ScreenLLM, a set of multimodal large language models (MLLMs) tailored for advanced UI understanding and action prediction. Extensive experiments on both open-source and proprietary models show that ScreenLLM accurately models user behavior and predicts actions. Our work lays the foundation for scalable, robust, and intelligent GUI agents that enhance user interaction in diverse software environments.

cs.CL↗

arXiv2025

GUI-Bee: Align GUI Action Grounding to Novel Environments via Autonomous Exploration

Graphical User Interface (GUI) action grounding is a critical step in GUI automation that maps language instructions to actionable elements on GUI screens. Most recent works of GUI action grounding leverage large GUI datasets to fine-tune MLLMs. However, the fine-tuning data always covers limited GUI environments, and we find the performance of the resulting model deteriorates in novel environments. We argue that the GUI grounding models should be further aligned to the novel environments to reveal their full potential, when the inference is known to involve novel environments, i.e., environments not used during the previous fine-tuning. To realize this, we first propose GUI-Bee, an MLLM-based autonomous agent, to collect high-quality, environment-specific data through exploration and then continuously fine-tune GUI grounding models with the collected data. Our agent leverages a novel Q-value-Incentive In-Context Reinforcement Learning (Q-ICRL) method to optimize exploration efficiency and data quality. Additionally, we introduce NovelScreenSpot, a benchmark for testing how well the data can help align GUI action grounding models to novel environments and demonstrate the effectiveness of data collected by GUI-Bee in the experiments. Furthermore, we conduct an ablation study to validate the Q-ICRL method in enhancing the efficiency of GUI-Bee. Project page: https://gui-bee.github.io

cs.CL↗

arXiv2024

Knowledge Editing for Large Language Model with Knowledge Neuronal Ensemble

As real-world knowledge is constantly evolving, ensuring the timeliness and accuracy of a model's knowledge is crucial. This has made knowledge editing in large language models increasingly important. However, existing knowledge editing methods face several challenges, including parameter localization coupling, imprecise localization, and a lack of dynamic interaction across layers. In this paper, we propose a novel knowledge editing method called Knowledge Neuronal Ensemble (KNE). A knowledge neuronal ensemble represents a group of neurons encoding specific knowledge, thus mitigating the issue of frequent parameter modification caused by coupling in parameter localization. The KNE method enhances the precision and accuracy of parameter localization by computing gradient attribution scores for each parameter at each layer. During the editing process, only the gradients and losses associated with the knowledge neuronal ensemble are computed, with error backpropagation performed accordingly, ensuring dynamic interaction and collaborative updates among parameters. Experimental results on three widely used knowledge editing datasets show that the KNE method significantly improves the accuracy of knowledge editing and achieves, or even exceeds, the performance of the best baseline methods in portability and locality metrics.

cs.CL↗

arXiv2024

In situ Qubit Frequency Tuning Circuit for Scalable Superconducting Quantum Computing: Scheme and Experiment

Frequency tunable qubit plays a significant role for scalable superconducting quantum processors. The state-of-the-art room-temperature electronics for tuning qubit frequency suffers from unscalable limit, such as heating problem, linear growth of control cables, etc. Here we propose a scalable scheme to tune the qubit frequency by using in situ superconducting circuit, which is based on radio frequency superconducting quantum interference device (rf-SQUID). We demonstrate both theoretically and experimentally that the qubit frequency could be modulated by inputting several single pulses into rf-SQUID. Compared with the traditional scheme, our scheme not only solves the heating problem, but also provides the potential to exponentially reduce the number of cables inside the dilute refrigerator and the room-temperature electronics resource for tuning qubit frequency, which is achieved by a time-division-multiplex (TDM) scheme combining rf-SQUID with switch arrays. With such TDM scheme, the number of cables could be reduced from the usual $\sim 3n$ to $\sim \log_2{(3n)} + 1$ for two-dimensional quantum processors comprising $n$ qubits and $\sim 2n$ couplers. Our work paves the way for large-scale control of superconducting quantum processor.

quant-ph↗

arXiv2024

HighlightMe: Detecting Highlights from Human-Centric Videos

We present a domain- and user-preference-agnostic approach to detect highlightable excerpts from human-centric videos. Our method works on the graph-based representation of multiple observable human-centric modalities in the videos, such as poses and faces. We use an autoencoder network equipped with spatial-temporal graph convolutions to detect human activities and interactions based on these modalities. We train our network to map the activity- and interaction-based latent structural representations of the different modalities to per-frame highlight scores based on the representativeness of the frames. We use these scores to compute which frames to highlight and stitch contiguous frames to produce the excerpts. We train our network on the large-scale AVA-Kinetics action dataset and evaluate it on four benchmark video highlight datasets: DSH, TVSum, PHD2, and SumMe. We observe a 4-12% improvement in the mean average precision of matching the human-annotated highlights over state-of-the-art methods in these datasets, without requiring any user-provided preferences or dataset-specific fine-tuning.

cs.CV↗

arXiv2024

Show Me What I Like: Detecting User-Specific Video Highlights Using Content-Based Multi-Head Attention

We propose a method to detect individualized highlights for users on given target videos based on their preferred highlight clips marked on previous videos they have watched. Our method explicitly leverages the contents of both the preferred clips and the target videos using pre-trained features for the objects and the human activities. We design a multi-head attention mechanism to adaptively weigh the preferred clips based on their object- and human-activity-based contents, and fuse them using these weights into a single feature representation for each user. We compute similarities between these per-user feature representations and the per-frame features computed from the desired target videos to estimate the user-specific highlight clips from the target videos. We test our method on a large-scale highlight detection dataset containing the annotated highlights of individual users. Compared to current baselines, we observe an absolute improvement of 2-4% in the mean average precision of the detected highlights. We also perform extensive ablation experiments on the number of preferred highlight clips associated with each user as well as on the object- and human-activity-based feature representations to validate that our method is indeed both content-based and user-specific.

cs.CV↗

arXiv2024

Improving Domain Generalization in Self-supervised Monocular Depth Estimation via Stabilized Adversarial Training

Learning a self-supervised Monocular Depth Estimation (MDE) model with great generalization remains significantly challenging. Despite the success of adversarial augmentation in the supervised learning generalization, naively incorporating it into self-supervised MDE models potentially causes over-regularization, suffering from severe performance degradation. In this paper, we conduct qualitative analysis and illuminate the main causes: (i) inherent sensitivity in the UNet-alike depth network and (ii) dual optimization conflict caused by over-regularization. To tackle these issues, we propose a general adversarial training framework, named Stabilized Conflict-optimization Adversarial Training (SCAT), integrating adversarial data augmentation into self-supervised MDE methods to achieve a balance between stability and generalization. Specifically, we devise an effective scaling depth network that tunes the coefficients of long skip connection and effectively stabilizes the training process. Then, we propose a conflict gradient surgery strategy, which progressively integrates the adversarial gradient and optimizes the model toward a conflict-free direction. Extensive experiments on five benchmarks demonstrate that SCAT can achieve state-of-the-art performance and significantly improve the generalization capability of existing self-supervised MDE methods.

cs.CV↗

arXiv2024

The ALMA-QUARKS Survey: Fibers' role in star formation unveiled in an intermediate-mass protocluster region of the Vela D cloud

In this paper, we present a detailed analysis of the IRS 17 filament within the intermediate-mass protocluster IRAS 08448-4343 (of $\sim\,10^3\,\rm L_{\odot}$), using ALMA data from the ATOMS 3-mm and QUARKS 1.3-mm surveys. The IRS 17 filament, which spans $\sim$54000 au ($0.26\,\rm pc$) in length and $\sim$4000 au ($0.02\,\rm pc$) in width, exhibits a complex, multi-component velocity field, and harbours hierarchical substructures. These substructures include three bundles of seven velocity-coherent fibers, and 29 dense ($n\sim 10^8\,\rm cm^{-3}$) condensations. The fibers have a median length of $\sim 4500\,\rm au$ and a median width of $\sim 1400\,\rm au$. Among these fibers, four are identified as ``fertile", each hosting at least three dense condensations, which are regarded as the ``seeds" of star formation. While the detected cores are randomly spaced within the IRS\,17 filament based on the 3-mm dust continuum image, periodic spacing ($\sim1600\,\rm au$) of condensations is observed in the fertile fibers according to the 1.3-mm dust map, consistent with the predictions of linear isothermal cylinder fragmentation models. These findings underscore the crucial role of fibers in star formation and suggest a hierarchical fragmentation process that extends from the filament to the fibers, and ultimately, to the smallest-scale condensations.

astro-ph.GA↗

arXiv2024

Transforming Image Super-Resolution: A ConvFormer-based Efficient Approach

Recent progress in single-image super-resolution (SISR) has achieved remarkable performance, yet the computational costs of these methods remain a challenge for deployment on resource-constrained devices. In particular, transformer-based methods, which leverage self-attention mechanisms, have led to significant breakthroughs but also introduce substantial computational costs. To tackle this issue, we introduce the Convolutional Transformer layer (ConvFormer) and propose a ConvFormer-based Super-Resolution network (CFSR), offering an effective and efficient solution for lightweight image super-resolution. The proposed method inherits the advantages of both convolution-based and transformer-based approaches. Specifically, CFSR utilizes large kernel convolutions as a feature mixer to replace the self-attention module, efficiently modeling long-range dependencies and extensive receptive fields with minimal computational overhead. Furthermore, we propose an edge-preserving feed-forward network (EFN) designed to achieve local feature aggregation while effectively preserving high-frequency information. Extensive experiments demonstrate that CFSR strikes an optimal balance between computational cost and performance compared to existing lightweight SR methods. When benchmarked against state-of-the-art methods such as ShuffleMixer, the proposed CFSR achieves a gain of 0.39 dB on the Urban100 dataset for the x2 super-resolution task while requiring 26\% and 31\% fewer parameters and FLOPs, respectively. The code and pre-trained models are available at https://github.com/Aitical/CFSR.

cs.CV↗

arXiv2024

LayoutDETR: Detection Transformer Is a Good Multimodal Layout Designer

Graphic layout designs play an essential role in visual communication. Yet handcrafting layout designs is skill-demanding, time-consuming, and non-scalable to batch production. Generative models emerge to make design automation scalable but it remains non-trivial to produce designs that comply with designers' multimodal desires, i.e., constrained by background images and driven by foreground content. We propose LayoutDETR that inherits the high quality and realism from generative modeling, while reformulating content-aware requirements as a detection problem: we learn to detect in a background image the reasonable locations, scales, and spatial relations for multimodal foreground elements in a layout. Our solution sets a new state-of-the-art performance for layout generation on public benchmarks and on our newly-curated ad banner dataset. We integrate our solution into a graphical system that facilitates user studies, and show that users prefer our designs over baselines by significant margins. Code, models, dataset, and demos are available at https://github.com/salesforce/LayoutDETR.

cs.CV↗

arXiv2024

The RoboDepth Challenge: Methods and Advancements Towards Robust Depth Estimation

Accurate depth estimation under out-of-distribution (OoD) scenarios, such as adverse weather conditions, sensor failure, and noise contamination, is desirable for safety-critical applications. Existing depth estimation systems, however, suffer inevitably from real-world corruptions and perturbations and are struggled to provide reliable depth predictions under such cases. In this paper, we summarize the winning solutions from the RoboDepth Challenge -- an academic competition designed to facilitate and advance robust OoD depth estimation. This challenge was developed based on the newly established KITTI-C and NYUDepth2-C benchmarks. We hosted two stand-alone tracks, with an emphasis on robust self-supervised and robust fully-supervised depth estimation, respectively. Out of more than two hundred participants, nine unique and top-performing solutions have appeared, with novel designs ranging from the following aspects: spatial- and frequency-domain augmentations, masked image modeling, image restoration and super-resolution, adversarial training, diffusion-based noise suppression, vision-language pre-training, learned model ensembling, and hierarchical feature enhancement. Extensive experimental analyses along with insightful observations are drawn to better understand the rationale behind each design. We hope this challenge could lay a solid foundation for future research on robust and reliable depth estimation and beyond. The datasets, competition toolkit, workshop recordings, and source code from the winning teams are publicly available on the challenge website.

cs.CV↗

arXiv2024

NH3 (1,1) hyperfine intensity anomalies in infall sources

Identifying infall motions is crucial for our understanding of accretion processes in regions of star formation. The NH3 (1,1) hyperfine intensity anomaly (HIA) has been proposed to be a readily usable tracer for such infall motions in star-forming regions harboring young stellar objects at very early evolutionary stages. In this paper, we seek to study the HIA toward fifteen infall candidate regions to assess its reliability as an infall tracer. By using deep observations of the NH3 (1,1) transition with the Effelsberg 100 m telescope, HIAs have been identified toward all the targets. Fourteen out of fifteen sources exhibit anomalous intensities either in the inner or outer satellite lines. All the derived HIAs conform to the framework of the existing two models, namely, hyperfine selective trapping (HST) and systematic contraction or expansion motion (CE) models. In our sample of infall candidates, a majority of the HIAs remain consistent with the HST model. Only in three targets, the HIAs are consistent with infall motions under the CE model. Thus HIAs could be used as an infall tracer but seem not highly sensitive to infall motions in our single-dish data. Nevertheless, the emission could be blended with emission from outflow activities. HIAs consistent with the HST model show stronger anomalies with increasing kinetic temperatures (Tk), which is expected by the HST model. On the other hand, HIAs consistent with infall motions show little dependence on Tk. Therefore, HIAs may preferably trace infall of cold gas.

astro-ph.GA↗

arXiv2024

Hybrid Cost Volume for Memory-Efficient Optical Flow

Current state-of-the-art flow methods are mostly based on dense all-pairs cost volumes. However, as image resolution increases, the computational and spatial complexity of constructing these cost volumes grows at a quartic rate, making these methods impractical for high-resolution images. In this paper, we propose a novel Hybrid Cost Volume for memory-efficient optical flow, named HCV. To construct HCV, we first propose a Top-k strategy to separate the 4D cost volume into two global 3D cost volumes. These volumes significantly reduce memory usage while retaining a substantial amount of matching information. We further introduce a local 4D cost volume with a local search space to supplement the local information for HCV. Based on HCV, we design a memory-efficient optical flow network, named HCVFlow. Compared to the recurrent flow methods based the all-pairs cost volumes, our HCVFlow significantly reduces memory consumption while ensuring high accuracy. We validate the effectiveness and efficiency of our method on the Sintel and KITTI datasets and real-world 4K (2160*3840) resolution images. Extensive experiments show that our HCVFlow has very low memory usage and outperforms other memory-efficient methods in terms of accuracy. The code is publicly available at https://github.com/gangweiX/HCVFlow.

cs.CV↗

arXiv2024

Kinematics and star formation of hub-filament systems in W49A

W49A is a prominent giant molecular cloud (GMC) that exhibits strong star formation activities, yet its structural and kinematic properties remain uncertain. Our study aims to investigate the large-scale structure and kinematics of W49A, and elucidate the role of filaments and hub-filament systems (HFSs) in its star formation activity. We utilized continuum data from Herschel and the James Clerk Maxwell Telescope (JCMT) as well as the molecular lines 12CO (3-2), 13CO (3-2), and C18O (3-2) to identify filaments and HFS structures within W49A. Further analysis focused on the physical properties, kinematics, and mass transport within these structures. Additionally, recombination line emission from the H I/OH/Recombination (THOR) line survey was employed to trace the central H II region and ionized gas. Our findings reveal that W49A comprises one blue-shifted (B-S) HFS and one red-shifted (R-S) HFS, each with multiple filaments and dense hubs. Notably, significant velocity gradients were detected along these filaments, indicative of material transport toward the hubs. High mass accretion rates along the filaments facilitate the formation of massive stars in the HFSs. Furthermore, the presence of V-shaped structures around clumps in position-velocity diagrams suggests ongoing gravitational collapse and local star formation within the filaments. Our results indicate that W49A consists of one R-S HFS and one B-S HFS, and that the material transport from filaments to the hub promotes the formation of massive stars in the hub. These findings underscore the significance of HFSs in shaping the star formation history of W49A.

astro-ph.GA↗

arXiv2024

Learning from Students: Applying t-Distributions to Explore Accurate and Efficient Formats for LLMs

The increasing size of large language models (LLMs) traditionally requires low-precision integer formats to meet strict latency and power demands. Yet recently, alternative formats such as Normal Float (NF4) have increased model accuracy at the cost of increased chip area. In this work, we first conduct a large-scale analysis of LLM weights and activations across 30 networks and conclude that most distributions follow a Student's t-distribution. We then derive a new theoretically optimal format, Student Float (SF4), that improves over NF4 across modern LLMs, for example increasing the average accuracy on LLaMA2-7B by 0.76% across tasks. Using this format as a high-accuracy reference, we then propose augmenting E2M1 with two variants of supernormal support for higher model accuracy. Finally, we explore the quality and efficiency frontier across 11 datatypes by evaluating their model accuracy and hardware complexity. We discover a Pareto curve composed of INT4, E2M1, and E2M1 with supernormal support, which offers a continuous tradeoff between model accuracy and chip area. For example, E2M1 with supernormal support increases the accuracy of Phi-2 by up to 2.19% with 1.22% area overhead, enabling more LLM-based applications to be run at four bits. The supporting code is hosted at https://github.com/cornell-zhang/llm-datatypes.

cs.LG↗

arXiv2024

Channel Reciprocity Based Attack Detection for Securing UWB Ranging by Autoencoder

A variety of ranging threats represented by Ghost Peak attack have raised concerns regarding the security performance of Ultra-Wide Band (UWB) systems with the finalization of the IEEE 802.15.4z standard. Based on channel reciprocity, this paper proposes a low complexity attack detection scheme that compares Channel Impulse Response (CIR) features of both ranging sides utilizing an autoencoder with the capability of data compression and feature extraction. Taking Ghost Peak attack as an example, this paper demonstrates the effectiveness, feasibility and generalizability of the proposed attack detection scheme through simulation and experimental validation. The proposed scheme achieves an attack detection success rate of over 99% and can be implemented in current systems at low cost.

cs.CR↗