arXiv subjects

Rui Zhang

Publications and source records attributed to Rui Zhang.

At least 631 records · Page 35Linked to original sources

arXiv2024

Visual Foundation Models Boost Cross-Modal Unsupervised Domain Adaptation for 3D Semantic Segmentation

Unsupervised domain adaptation (UDA) is vital for alleviating the workload of labeling 3D point cloud data and mitigating the absence of labels when facing a newly defined domain. Various methods of utilizing images to enhance the performance of cross-domain 3D segmentation have recently emerged. However, the pseudo labels, which are generated from models trained on the source domain and provide additional supervised signals for the unseen domain, are inadequate when utilized for 3D segmentation due to their inherent noisiness and consequently restrict the accuracy of neural networks. With the advent of 2D visual foundation models (VFMs) and their abundant knowledge prior, we propose a novel pipeline VFMSeg to further enhance the cross-modal unsupervised domain adaptation framework by leveraging these models. In this work, we study how to harness the knowledge priors learned by VFMs to produce more accurate labels for unlabeled target domains and improve overall performance. We first utilize a multi-modal VFM, which is pre-trained on large scale image-text pairs, to provide supervised labels (VFM-PL) for images and point clouds from the target domain. Then, another VFM trained on fine-grained 2D masks is adopted to guide the generation of semantically augmented images and point clouds to enhance the performance of neural networks, which mix the data from source and target domains like view frustums (FrustumMixing). Finally, we merge class-wise prediction across modalities to produce more accurate annotations for unlabeled target domains. Our method is evaluated on various autonomous driving datasets and the results demonstrate a significant improvement for 3D segmentation task.

cs.CV↗

arXiv2024

Generation is better than Modification: Combating High Class Homophily Variance in Graph Anomaly Detection

Graph-based anomaly detection is currently an important research topic in the field of graph neural networks (GNNs). We find that in graph anomaly detection, the homophily distribution differences between different classes are significantly greater than those in homophilic and heterophilic graphs. For the first time, we introduce a new metric called Class Homophily Variance, which quantitatively describes this phenomenon. To mitigate its impact, we propose a novel GNN model named Homophily Edge Generation Graph Neural Network (HedGe). Previous works typically focused on pruning, selecting or connecting on original relationships, and we refer to these methods as modifications. Different from these works, our method emphasizes generating new relationships with low class homophily variance, using the original relationships as an auxiliary. HedGe samples homophily adjacency matrices from scratch using a self-attention mechanism, and leverages nodes that are relevant in the feature space but not directly connected in the original graph. Additionally, we modify the loss function to punish the generation of unnecessary heterophilic edges by the model. Extensive comparison experiments demonstrate that HedGe achieved the best performance across multiple benchmark datasets, including anomaly detection and edgeless node classification. The proposed model also improves the robustness under the novel Heterophily Attack with increased class homophily variance on other graph classification tasks.

cs.LG↗

arXiv2024

Insights from LHAASO and IceCube into the origin of the Galactic diffuse TeV--PeV emission

The high-energy diffuse gamma-ray emission and neutrino emission are expected from the Galactic plane, generated by hadronuclear interactions between cosmic rays (CR) and interstellar medium (ISM). Therefore, measurements of these diffuse emissions will provide important clues on the origin and nature of Galactic CRs. Comparing the latest observations of LHAASO and IceCube on the diffuse Galactic gamma-ray and neutrino emissions respectively, we suggest that the diffuse gamma-ray emission at multi-TeV energies contains a considerable contribution of a leptonic component. By modelling the gamma-ray halos powered by middle-aged pulsars in our Galaxy with taking into account the magnetic field configuration and the interstellar radiation field in the Galaxy, we demonstrate that the collective contribution of pulsar halos can account for the excess in the measured diffuse gamma-ray emission with respect to the predicted flux from CR-ISM interactions. The resulting one-dimensional profile along the Galactic longitude is also consistent with the observation.

astro-ph.HE↗

arXiv2024

TCP:Textual-based Class-aware Prompt tuning for Visual-Language Model

Prompt tuning represents a valuable technique for adapting pre-trained visual-language models (VLM) to various downstream tasks. Recent advancements in CoOp-based methods propose a set of learnable domain-shared or image-conditional textual tokens to facilitate the generation of task-specific textual classifiers. However, those textual tokens have a limited generalization ability regarding unseen domains, as they cannot dynamically adjust to the distribution of testing classes. To tackle this issue, we present a novel Textual-based Class-aware Prompt tuning(TCP) that explicitly incorporates prior knowledge about classes to enhance their discriminability. The critical concept of TCP involves leveraging Textual Knowledge Embedding (TKE) to map the high generalizability of class-level textual knowledge into class-aware textual tokens. By seamlessly integrating these class-aware prompts into the Text Encoder, a dynamic class-aware classifier is generated to enhance discriminability for unseen domains. During inference, TKE dynamically generates class-aware prompts related to the unseen classes. Comprehensive evaluations demonstrate that TKE serves as a plug-and-play module effortlessly combinable with existing methods. Furthermore, TCP consistently achieves superior performance while demanding less training time. Code:https://github.com/htyao89/Textual-based_Class-aware_prompt_tuning/

cs.CV↗

arXiv2024

Enhancing the hunt for new phenomena in dijet final-states using anomaly detection filters at the High-Luminosity Large Hadron Collider

In the realm of dijet searches in high-energy physics, a significant challenge has emerged: with experiments producing more and more data, the traditional methods of using analytic functions to describe dijet mass spectra start to fail. To address this, we suggest the application of an anomaly detection approach to eliminate less interesting background events based on event final states. This method not only bypasses the limitations of conventional background models but also significantly enhances our ability to detect potential signals of new physics. Through simulations that mimic the conditions of the upcoming High-Luminosity Large Hadron Collider, we demonstrate the strength and efficiency of this approach in dealing with large data volumes. The integration of unsupervised machine learning into our experimental framework paves the way for a promising avenue to unveil hidden physics discoveries within the overwhelming influx of data.

hep-ex↗

arXiv2024

Origin and Customization of Bandgap in Chiral Phononic Crystals

The wave equation governing the wave propagation in chiral phononic crystals, established through force equilibrium law, conceals the underlying physical information. This has led to a controversy over the bandgap mechanism. In this letter, we theoretically unveil the reason of this controversy, and put forward an alternative approach from wave behavior to formulate the wave equation, offering a new pathway to articulate the bandgap physics directly. We identify the obstacles in coupled acoustic and optic branches to widen and lower the bandgap, and introduce an approach based on spherical hinges to decrease the barriers, for customizing the bandgap frequency and width. Finally, we validate our proposal through numerical simulation and experimental demonstration.

physics.app-ph↗

arXiv2024

Deciphering and integrating invariants for neural operator learning with various physical mechanisms

Neural operators have been explored as surrogate models for simulating physical systems to overcome the limitations of traditional partial differential equation (PDE) solvers. However, most existing operator learning methods assume that the data originate from a single physical mechanism, limiting their applicability and performance in more realistic scenarios. To this end, we propose Physical Invariant Attention Neural Operator (PIANO) to decipher and integrate the physical invariants (PI) for operator learning from the PDE series with various physical mechanisms. PIANO employs self-supervised learning to extract physical knowledge and attention mechanisms to integrate them into dynamic convolutional layers. Compared to existing techniques, PIANO can reduce the relative error by 13.6\%-82.2\% on PDE forecasting tasks across varying coefficients, forces, or boundary conditions. Additionally, varied downstream tasks reveal that the PI embeddings deciphered by PIANO align well with the underlying invariants in the PDE systems, verifying the physical significance of PIANO. The source code will be publicly available at: https://github.com/optray/PIANO.

cs.LG↗

arXiv2024

A Tutorial on Environment-Aware Communications via Channel Knowledge Map for 6G

Sixth-generation (6G) mobile communication networks are expected to have dense infrastructures, large antenna size, wide bandwidth, cost-effective hardware, diversified positioning methods, and enhanced intelligence. Such trends bring both new challenges and opportunities for the practical design of 6G. On one hand, acquiring channel state information (CSI) in real time for all wireless links becomes quite challenging in 6G. On the other hand, there would be numerous data sources in 6G containing high-quality location-tagged channel data, e.g., the estimated channels or beams between base station (BS) and user equipment (UE), making it possible to better learn the local wireless environment. By exploiting this new opportunity and for tackling the CSI acquisition challenge, there is a promising paradigm shift from the conventional environment-unaware communications to the new environment-aware communications based on the novel approach of channel knowledge map (CKM). This article aims to provide a comprehensive overview on environment-aware communications enabled by CKM to fully harness its benefits for 6G. First, the basic concept of CKM is presented, followed by the comparison of CKM with various existing channel inference techniques. Next, the main techniques for CKM construction are discussed, including both environment model-free and environment model-assisted approaches. Furthermore, a general framework is presented for the utilization of CKM to achieve environment-aware communications, followed by some typical CKM-aided communication scenarios. Finally, important open problems in CKM research are highlighted and potential solutions are discussed to inspire future work.

cs.IT↗

arXiv2024

MT-Ranker: Reference-free machine translation evaluation by inter-system ranking

Traditionally, Machine Translation (MT) Evaluation has been treated as a regression problem -- producing an absolute translation-quality score. This approach has two limitations: i) the scores lack interpretability, and human annotators struggle with giving consistent scores; ii) most scoring methods are based on (reference, translation) pairs, limiting their applicability in real-world scenarios where references are absent. In practice, we often care about whether a new MT system is better or worse than some competitors. In addition, reference-free MT evaluation is increasingly practical and necessary. Unfortunately, these two practical considerations have yet to be jointly explored. In this work, we formulate the reference-free MT evaluation into a pairwise ranking problem. Given the source sentence and a pair of translations, our system predicts which translation is better. In addition to proposing this new formulation, we further show that this new paradigm can demonstrate superior correlation with human judgments by merely using indirect supervision from natural language inference and weak supervision from our synthetic data. In the context of reference-free evaluation, MT-Ranker, trained without any human annotations, achieves state-of-the-art results on the WMT Shared Metrics Task benchmarks DARR20, MQM20, and MQM21. On a more challenging benchmark, ACES, which contains fine-grained evaluation criteria such as addition, omission, and mistranslation errors, MT-Ranker marks state-of-the-art against reference-free as well as reference-based baselines.

cs.CL↗

arXiv2024

IRS Aided Millimeter-Wave Sensing and Communication: Beam Scanning, Beam Splitting, and Performance Analysis

Integrated sensing and communication (ISAC) has attracted growing interests for enabling the future 6G wireless networks, due to its capability of sharing spectrum and hardware resources between communication and sensing systems. However, existing works on ISAC usually need to modify the communication protocol to cater for the new sensing performance requirement, which may be difficult to implement in practice. In this paper, we study a new intelligent reflecting surface (IRS) aided millimeter-wave (mmWave) ISAC system by exploiting the distinct beam scanning operation in mmWave communications to achieve efficient sensing at the same time. First, we propose a two-phase ISAC protocol aided by a semi-passive IRS, consisting of beam scanning and data transmission. Specifically, in the beam scanning phase, the IRS finds the optimal beam for reflecting signals from the base station to a communication user via its passive elements. Meanwhile, the IRS directly estimates the angle of a nearby target based on echo signals from the target using its equipped active sensing element. Then, in the data transmission phase, the sensing accuracy is further improved by leveraging the data signals via possible IRS beam splitting. Next, we derive the achievable rate of the communication user as well as the Cramér-Rao bound and the approximate mean square error of the target angle estimation Finally, extensive simulation results are provided to verify our analysis as well as the effectiveness of the proposed scheme.

cs.IT↗

arXiv2024

Towards Better Understanding of User Satisfaction in Open-Domain Conversational Search

With the increasing popularity of conversational search, how to evaluate the performance of conversational search systems has become an important question in the IR community. Existing works on conversational search evaluation can mainly be categorized into two streams: (1) constructing metrics based on semantic similarity (e.g. BLUE, METEOR and BERTScore), or (2) directly evaluating the response ranking performance of the system using traditional search methods (e.g. nDCG, RBP and nERR). However, these methods either ignore the information need of the user or ignore the mixed-initiative property of conversational search. This raises the question of how to accurately model user satisfaction in conversational search scenarios. Since explicitly asking users to provide satisfaction feedback is difficult, traditional IR studies often rely on the Cranfield paradigm (i.e., third-party annotation) and user behavior modeling to estimate user satisfaction in search. However, the feasibility and effectiveness of these two approaches have not been fully explored in conversational search. In this paper, we dive into the evaluation of conversational search from the perspective of user satisfaction. We build a novel conversational search experimental platform and construct a Chinese open-domain conversational search behavior dataset containing rich annotations and search behavior data. We also collect third-party satisfaction annotation at the session-level and turn-level, to investigate the feasibility of the Cranfield paradigm in the conversational search scenario. Experimental results show both some consistency and considerable differences between the user satisfaction annotations and third-party annotations. We also propose dialog continuation or ending behavior models (DCEBM) to capture session-level user satisfaction based on turn-level information.

cs.IR↗

arXiv2024

EL-VIT: Probing Vision Transformer with Interactive Visualization

Nowadays, Vision Transformer (ViT) is widely utilized in various computer vision tasks, owing to its unique self-attention mechanism. However, the model architecture of ViT is complex and often challenging to comprehend, leading to a steep learning curve. ViT developers and users frequently encounter difficulties in interpreting its inner workings. Therefore, a visualization system is needed to assist ViT users in understanding its functionality. This paper introduces EL-VIT, an interactive visual analytics system designed to probe the Vision Transformer and facilitate a better understanding of its operations. The system consists of four layers of visualization views. The first three layers include model overview, knowledge background graph, and model detail view. These three layers elucidate the operation process of ViT from three perspectives: the overall model architecture, detailed explanation, and mathematical operations, enabling users to understand the underlying principles and the transition process between layers. The fourth interpretation view helps ViT users and experts gain a deeper understanding by calculating the cosine similarity between patches. Our two usage scenarios demonstrate the effectiveness and usability of EL-VIT in helping ViT users understand the working mechanism of ViT.

cs.AI↗

arXiv2024

Assessing and Understanding Creativity in Large Language Models

In the field of natural language processing, the rapid development of large language model (LLM) has attracted more and more attention. LLMs have shown a high level of creativity in various tasks, but the methods for assessing such creativity are inadequate. The assessment of LLM creativity needs to consider differences from humans, requiring multi-dimensional measurement while balancing accuracy and efficiency. This paper aims to establish an efficient framework for assessing the level of creativity in LLMs. By adapting the modified Torrance Tests of Creative Thinking, the research evaluates the creative performance of various LLMs across 7 tasks, emphasizing 4 criteria including Fluency, Flexibility, Originality, and Elaboration. In this context, we develop a comprehensive dataset of 700 questions for testing and an LLM-based evaluation method. In addition, this study presents a novel analysis of LLMs' responses to diverse prompts and role-play situations. We found that the creativity of LLMs primarily falls short in originality, while excelling in elaboration. Besides, the use of prompts and the role-play settings of the model significantly influence creativity. Additionally, the experimental results also indicate that collaboration among multiple LLMs can enhance originality. Notably, our findings reveal a consensus between human evaluations and LLMs regarding the personality traits that influence creativity. The findings underscore the significant impact of LLM design on creativity and bridges artificial intelligence and human creativity, offering insights into LLMs' creativity and potential applications.

cs.CL↗

arXiv2024

Astronomical Knowledge Entity Extraction in Astrophysics Journal Articles via Large Language Models

Astronomical knowledge entities, such as celestial object identifiers, are crucial for literature retrieval and knowledge graph construction, and other research and applications in the field of astronomy. Traditional methods of extracting knowledge entities from texts face challenges like high manual effort, poor generalization, and costly maintenance. Consequently, there is a pressing need for improved methods to efficiently extract them. This study explores the potential of pre-trained Large Language Models (LLMs) to perform astronomical knowledge entity extraction (KEE) task from astrophysical journal articles using prompts. We propose a prompting strategy called Prompt-KEE, which includes five prompt elements, and design eight combination prompts based on them. Celestial object identifier and telescope name, two most typical astronomical knowledge entities, are selected to be experimental object. And we introduce four currently representative LLMs, namely Llama-2-70B, GPT-3.5, GPT-4, and Claude 2. To accommodate their token limitations, we construct two datasets: the full texts and paragraph collections of 30 articles. Leveraging the eight prompts, we test on full texts with GPT-4 and Claude 2, on paragraph collections with all LLMs. The experimental results demonstrated that pre-trained LLMs have the significant potential to perform KEE tasks in astrophysics journal articles, but there are differences in their performance. Furthermore, we analyze some important factors that influence the performance of LLMs in entity extraction and provide insights for future KEE tasks in astrophysical articles using LLMs.

astro-ph.IM↗

arXiv2024

Performance Analysis and Optimization for Movable Antenna Aided Wideband Communications

Movable antenna (MA) has emerged as a promising technology to enhance wireless communication performance by enabling the local movement of antennas at the transmitter (Tx) and/or receiver (Rx) for achieving more favorable channel conditions. As the existing studies on MA-aided wireless communications have mainly considered narrow-band transmission in flat fading channels, we investigate in this paper the MA-aided wideband communications employing orthogonal frequency division multiplexing (OFDM) in frequency-selective fading channels. Under the general multi-tap field-response channel model, the wireless channel variations in both space and frequency are characterized with different positions of the MAs. Unlike the narrow-band transmission where the optimal MA position at the Tx/Rx simply maximizes the single-tap channel amplitude, the MA position in the wideband case needs to balance the amplitudes and phases over multiple channel taps in order to maximize the OFDM transmission rate over multiple frequency subcarriers. First, we derive an upper bound on the OFDM achievable rate in closed form when the size of the Tx/Rx region for antenna movement is arbitrarily large. Next, we develop a parallel greedy ascent (PGA) algorithm to obtain locally optimal solutions to the MAs' positions for OFDM rate maximization subject to finite-size Tx/Rx regions. To reduce computational complexity, a simplified PGA algorithm is also provided to optimize the MAs' positions more efficiently. Simulation results demonstrate that the proposed PGA algorithms can approach the OFDM rate upper bound closely with the increase of Tx/Rx region sizes and outperform conventional systems with fixed-position antennas (FPAs) under the wideband channel setup.

cs.IT↗

arXiv2024

GMLake: Efficient and Transparent GPU Memory Defragmentation for Large-scale DNN Training with Virtual Memory Stitching

Large-scale deep neural networks (DNNs), such as large language models (LLMs), have revolutionized the artificial intelligence (AI) field and become increasingly popular. However, training or fine-tuning such models requires substantial computational power and resources, where the memory capacity of a single acceleration device like a GPU is one of the most important bottlenecks. Owing to the prohibitively large overhead (e.g., $10 \times$) of GPUs' native memory allocator, DNN frameworks like PyTorch and TensorFlow adopt a caching allocator that maintains a memory pool with a splitting mechanism for fast memory (de)allocation. Unfortunately, the caching allocator's efficiency degrades quickly for popular memory reduction techniques such as recomputation, offloading, distributed training, and low-rank adaptation. The primary reason is that those memory reduction techniques introduce frequent and irregular memory (de)allocation requests, leading to severe fragmentation problems for the splitting-based caching allocator. To mitigate this fragmentation problem, we propose a novel memory allocation framework based on low-level GPU virtual memory management called GPU memory lake (GMLake). GMLake employs a novel virtual memory stitching (VMS) mechanism, which can fuse or combine non-contiguous memory blocks with a virtual memory address mapping. GMLake can reduce an average of 9.2 GB (up to 25 GB) GPU memory usage and 15% (up to 33% ) fragmentation among eight LLM models on GPU A100 with 80 GB memory. GMLake is completely transparent to the DNN models and memory reduction techniques and ensures the seamless execution of resource-intensive deep-learning tasks. We have open-sourced GMLake at https://github.com/intelligent-machine-learning/glake/tree/main/GMLake.

cs.DC↗

arXiv2024

Passive Reflection Codebook Design for IRS-Integrated Access Point

Intelligent reflecting surface (IRS) has emerged as a promising technique to extend the wireless signal coverage of access point (AP) and improve the communication performance cost-effectively. In order to reduce the path-loss of the cascaded user-IRS-AP channels, the IRS-integrated AP architecture has been proposed to deploy the IRSs and the antenna array of the AP within the same antenna radome. To reduce the pilot overhead for estimating all IRS-involved channels, in this paper, we propose a novel codebook-based IRS reflection design for the IRS-integrated AP to enhance the coverage performance in a given area. In particular, the codebook consisting of a small number of codewords is designed offline by employing an efficient sector division strategy based on the azimuth angle. To ensure the performance of each sector, we optimize its corresponding codeword for IRS reflection pattern to maximize the sector-min-average-effective-channel-power (SMAECP) by applying the alternating optimization (AO) and semidefinite relaxation (SDR) methods. With the designed codebook, the AP performs the IRS reflection training by sequentially applying all codewords and selects the one achieving the best communication performance for data transmission. Numerical results show that our proposed codebook design can enhance the average channel power of the whole coverage area, as compared to the system without IRS. Moreover, our proposed codebook-based IRS reflection design is shown to achieve significant performance gain over other benchmark schemes in both single-user and multi-user transmissions.

cs.IT↗

arXiv2024

Machine Eye for Defects: Machine Learning-Based Solution to Identify and Characterize Topological Defects in Textured Images of Nematic Materials

Topological defects play a key role in the structures and dynamics of liquid crystals (LCs) and other ordered systems. There is a recent interest in studying defects in different biological systems with distinct textures. However, a robust method to directly recognize defects and extract their structural features from various traditional and nontraditional nematic systems remains challenging to date. Here we present a machine learning solution, termed Machine Eye for Defects (MED), for automated defect analysis in images with diverse nematic textures. MED seamlessly integrates state-of-the-art object detection networks, Segment Anything Model, and vision transformer algorithms with tailored computer vision techniques. We show that MED can accurately identify the positions, winding numbers, and orientations of $\pm 1/2$ defects across distinct cellular contours, sparse vector fields of nematic directors, actin filaments, microtubules, and simulation images of Gay--Berne particles. MED performs faster than conventional defect detection method and can achieve over 90\% accuracy on recognizing $\pm1/2$ defects and their orientations from vector fields and experimental tissue images. We further demonstrate that MED can identify defect types that are not included in the training data, such as giant-core defects and defects with higher winding number. Remarkably, MED provides correct structural information about $\pm 1$ defects, i.e., the phase angle for $+1$ defects and the orientation angle for $-1$ defects. As such, MED stands poised to transform studies of diverse ordered systems by providing automated, rapid, accurate, and insightful defect analysis.

cond-mat.soft↗