arXiv subjects

Yongqing Wang

Publications and source records attributed to Yongqing Wang.

At least 19 recordsLinked to original sources

arXiv2026

MCPHunt: An Evaluation Framework for Cross-Boundary Data Propagation in Multi-Server MCP Agents

Multi-server MCP agents create an information-flow control problem: faithful tool composition can turn individually benign read/write permissions into cross-boundary credential propagation -- a structural side effect of workflow topology, not necessarily malicious model behavior. We present MCPHunt, to our knowledge the first controlled benchmark that isolates non-adversarial, verbatim credential propagation across multi-server MCP trust boundaries, with three methodological contributions: (1) canary-based taint tracking that reduces propagation detection to objective string matching; (2) an environment-controlled coverage design with risky, benign, and hard-negative conditions that validates pipeline soundness and controls for credential-format confounds; (3) CRS stratification that disentangles task-mandated propagation (faithful execution of verbatim-transfer instructions) from policy-violating propagation (credentials included despite the option to redact). Across 3,615 main-benchmark traces from 5 models spanning 147 tasks and 9 mechanism families, policy-violating propagation rates reach 11.5--41.3% across all models. This propagation is pathway-specific (25x cross-mechanism range) and concentrated in browser-mediated data flows; hard-negative controls provide evidence that production-format credentials are not necessary -- prompt-directed cross-boundary data flow is sufficient. A prompt-mitigation study across 3 models reduces policy-violating propagation by up to 97% while preserving 80.5% utility, but effectiveness varies with instruction-following capability -- suggesting that prompt-level defenses alone may not suffice. Code, traces, and labeling pipeline are released under MIT and CC BY 4.0.

cs.AI↗

arXiv2026

GLAP: General contrastive audio-text pretraining across domains and languages

Contrastive Language Audio Pretraining (CLAP) is a widely-used method to bridge the gap between audio and text domains. Current CLAP methods enable sound and music retrieval in English, ignoring multilingual spoken content. To address this, we introduce general language audio pretraining (GLAP), which expands CLAP with multilingual and multi-domain abilities. GLAP demonstrates its versatility by achieving competitive performance on standard audio-text retrieval benchmarks like Clotho and AudioCaps, while significantly surpassing existing methods in speech retrieval and classification tasks. Additionally, GLAP achieves strong results on widely used sound-event zero-shot benchmarks, while simultaneously outperforming previous methods on speech content benchmarks. Further keyword spotting evaluations across 50 languages emphasize GLAP's advanced multilingual capabilities. Finally, multilingual sound and music understanding is evaluated across four languages. Checkpoints and Source: https://github.com/xiaomi-research/dasheng-glap.

cs.SD↗

arXiv2025

RIS- and Multi-Snapshot-Enabled SISO 3D Position and Velocity Estimation With Single Base Station

Reconfigurable intelligent surface (RIS) panels can act as cost-effective anchors for radio localization, complementing conventional base station (BS) anchors. This paper investigates joint three-dimensional position and velocity estimation (3D-JPVE) in single-input single-output (SISO) systems with only one BS available. We first theoretically show that 3D-JPVE is infeasible when relying solely on a single RIS or on multiple snapshots alone. To address this, we propose combining RIS deployment with multi-snapshot utilization to enable realizable 3D-JPVE. A two-stage method is developed for multi-snapshot channel parameter estimation, comprising a tensor-based coarse estimation step followed by a maximum likelihood refinement step. In particular, we introduce a third-order tensor formulation to decompose the challenging 3D joint angle-of-departure and Doppler shift estimation (3D-JADE) into two tractable subproblems, which are jointly solved via a low-complexity alternating optimization approach. Building on the channel parameter estimates, we further design a two-stage low-complexity method for optimal 3D-JPVE: coarse estimation is obtained from differential measurements through linear equations, and the preliminary results are refined iteratively using the original measurements. Moreover, we derive the closed-form Cramer-Rao lower bound (CRLB) and show that the proposed 3D-JPVE method approaches CRLB-level accuracy. Simulation results confirm the statistical efficiency of the proposed estimators and demonstrate substantial 3D-JPVE performance gains when deploying active RIS compared to passive RIS.

cs.IT↗

arXiv2025

BotTrans: A Multi-Source Graph Domain Adaptation Approach for Social Bot Detection

Transferring extensive knowledge from relevant social networks has emerged as a promising solution to overcome label scarcity in detecting social bots and other anomalies with GNN-based models. However, effective transfer faces two critical challenges. Firstly, the network heterophily problem, which is caused by bots hiding malicious behaviors via indiscriminately interacting with human users, hinders the model's ability to learn sufficient and accurate bot-related knowledge from source domains. Secondly, single-source transfer might lead to inferior and unstable results, as the source network may embody weak relevance to the task and provide limited knowledge. To address these challenges, we explore multiple source domains and propose a multi-source graph domain adaptation model named \textit{BotTrans}. We initially leverage the labeling knowledge shared across multiple source networks to establish a cross-source-domain topology with increased network homophily. We then aggregate cross-domain neighbor information to enhance the discriminability of source node embeddings. Subsequently, we integrate the relevance between each source-target pair with model optimization, which facilitates knowledge transfer from source networks that are more relevant to the detection task. Additionally, we propose a refinement strategy to improve detection performance by utilizing semantic knowledge within the target domain. Extensive experiments on real-world datasets demonstrate that \textit{BotTrans} outperforms the existing state-of-the-art methods, revealing its efficacy in leveraging multi-source knowledge when the target detection task is unlabeled.

cs.AI↗

arXiv2025

StarVC: A Unified Auto-Regressive Framework for Joint Text and Speech Generation in Voice Conversion

Voice Conversion (VC) modifies speech to match a target speaker while preserving linguistic content. Traditional methods usually extract speaker information directly from speech while neglecting the explicit utilization of linguistic content. Since VC fundamentally involves disentangling speaker identity from linguistic content, leveraging structured semantic features could enhance conversion performance. However, previous attempts to incorporate semantic features into VC have shown limited effectiveness, motivating the integration of explicit text modeling. We propose StarVC, a unified autoregressive VC framework that first predicts text tokens before synthesizing acoustic features. The experiments demonstrate that StarVC outperforms conventional VC methods in preserving both linguistic content (i.e., WER and CER) and speaker characteristics (i.e., SECS and MOS). Audio demo can be found at: https://thuhcsi.github.io/StarVC/.

cs.MM↗

arXiv2025

Structured Tensor Decomposition Based Channel Estimation and Double Refinements for Active RIS Empowered Broadband Systems

Channel parameter recovery is critical for the next-generation reconfigurable intelligent surface (RIS)-empowered communications and sensing. Tensor-based mechanisms are particularly effective, inherently capturing the multi-dimensional nature of wireless channels. However, existing studies assume either a line-of-sight (LOS) scenario or a blocked TX-RX channel. This paper solves a novel problem: tensor-based channel parameter estimation for active RIS-aided multiple-antenna broadband connections in fully multipath environments with the TX-RX link. System settings are customized to construct a fifth-order canonical polyadic (CP) signal tensor that matches the five-dimensional channel. Four tensor factors contain redundant columns, rendering the classical Kruskal's condition for decomposition uniqueness unsatisfied. The fifth-order Vandermonde structured CP decomposition (VSCPD) is developed to address this challenge, making the tensor factorization problem solvable using only linear algebra and offering a relaxed general uniqueness condition. With VSCPD as a perfect decoupling scheme, a sequential triple-stage channel estimation algorithm is proposed based on one-dimensional parameter estimation. The first stage enables multipath identification and algebraic coarse estimation. The following two stages offer optional successive refinements at the cost of increased complexity. The closed-form Cramer-Rao lower bound (CRLB) is derived to assess the estimation performance. Herein, the noise covariance matrix depends on multipath parameters in our active-RIS scenario. Numerical results are provided to verify the effectiveness of proposed algorithms under various evaluation metrics. Our results also show that active RIS can significantly improve channel estimation performance compared to passive RIS.

cs.IT↗

arXiv2024

FCS-HGNN: Flexible Multi-type Community Search in Heterogeneous Information Networks

Community search is a personalized community discovery problem designed to identify densely connected subgraphs containing the query node. Recently, community search in heterogeneous information networks (HINs) has received considerable attention. Existing methods typically focus on modeling relationships in HINs through predefined meta-paths or user-specified relational constraints. However, metapath-based methods are primarily designed to identify single-type communities with nodes of the same type rather than multi-type communities involving nodes of different types. Constraint-based methods require users to have a good understanding of community patterns to define a suitable set of relational constraints, which increases the burden on users. In this paper, we propose FCS-HGNN, a novel method for flexibly identifying both single-type and multi-type communities in HINs. Specifically, FCS-HGNN extracts complementary information from different views and dynamically considers the contribution of each relation instead of treating them equally, thereby capturing more fine-grained heterogeneous information. Furthermore, to improve efficiency on large-scale graphs, we further propose LS-FCS-HGNN, which incorporates i) the neighbor sampling strategy to improve training efficiency, and ii) the depth-based heuristic search strategy to improve query efficiency. We conducted extensive experiments to demonstrate the superiority of our proposed methods over state-of-the-art methods, achieving average improvements of 14.3% and 11.1% on single-type and multi-type communities, respectively.

cs.SI↗

arXiv2024

Enhancing Automated Audio Captioning via Large Language Models with Optimized Audio Encoding

Automated audio captioning (AAC) is an audio-to-text task to describe audio contents in natural language. Recently, the advancements in large language models (LLMs), with improvements in training approaches for audio encoders, have opened up possibilities for improving AAC. Thus, we explore enhancing AAC from three aspects: 1) a pre-trained audio encoder via consistent ensemble distillation (CED) is used to improve the effectivity of acoustic tokens, with a querying transformer (Q-Former) bridging the modality gap to LLM and compress acoustic tokens; 2) we investigate the advantages of using a Llama 2 with 7B parameters as the decoder; 3) another pre-trained LLM corrects text errors caused by insufficient training data and annotation ambiguities. Both the audio encoder and text decoder are optimized by low-rank adaptation (LoRA). Experiments show that each of these enhancements is effective. Our method obtains a 33.0 SPIDEr-FL score, outperforming the winner of DCASE 2023 Task 6A.

cs.SD↗

arXiv2024

Bridging Language Gaps in Audio-Text Retrieval

Audio-text retrieval is a challenging task, requiring the search for an audio clip or a text caption within a database. The predominant focus of existing research on English descriptions poses a limitation on the applicability of such models, given the abundance of non-English content in real-world data. To address these linguistic disparities, we propose a language enhancement (LE), using a multilingual text encoder (SONAR) to encode the text data with language-specific information. Additionally, we optimize the audio encoder through the application of consistent ensemble distillation (CED), enhancing support for variable-length audio-text retrieval. Our methodology excels in English audio-text retrieval, demonstrating state-of-the-art (SOTA) performance on commonly used datasets such as AudioCaps and Clotho. Simultaneously, the approach exhibits proficiency in retrieving content in seven other languages with only 10% of additional language-enhanced training data, yielding promising results. The source code is publicly available https://github.com/zyyan4/ml-clap.

cs.SD↗

arXiv2024

Scaling up masked audio encoder learning for general audio classification

Despite progress in audio classification, a generalization gap remains between speech and other sound domains, such as environmental sounds and music. Models trained for speech tasks often fail to perform well on environmental or musical audio tasks, and vice versa. While self-supervised (SSL) audio representations offer an alternative, there has been limited exploration of scaling both model and dataset sizes for SSL-based general audio classification. We introduce Dasheng, a simple SSL audio encoder, based on the efficient masked autoencoder framework. Trained with 1.2 billion parameters on 272,356 hours of diverse audio, Dasheng obtains significant performance gains on the HEAR benchmark. It outperforms previous works on CREMA-D, LibriCount, Speech Commands, VoxLingua, and competes well in music and environment classification. Dasheng features inherently contain rich speech, music, and environmental information, as shown in nearest-neighbor classification experiments. Code is available https://github.com/richermans/dasheng/.

cs.SD↗

arXiv2024

Streaming Audio Transformers for Online Audio Tagging

Transformers have emerged as a prominent model framework for audio tagging (AT), boasting state-of-the-art (SOTA) performance on the widely-used Audioset dataset. However, their impressive performance often comes at the cost of high memory usage, slow inference speed, and considerable model delay, rendering them impractical for real-world AT applications. In this study, we introduce streaming audio transformers (SAT) that combine the vision transformer (ViT) architecture with Transformer-Xl-like chunk processing, enabling efficient processing of long-range audio signals. Our proposed SAT is benchmarked against other transformer-based SOTA methods, achieving significant improvements in terms of mean average precision (mAP) at a delay of 2s and 1s, while also exhibiting significantly lower memory usage and computational overhead. Checkpoints are publicly available https://github.com/RicherMans/SAT.

cs.SD↗

arXiv2024

Graph Domain Adaptation: Challenges, Progress and Prospects

As graph representation learning often suffers from label scarcity problems in real-world applications, researchers have proposed graph domain adaptation (GDA) as an effective knowledge-transfer paradigm across graphs. In particular, to enhance model performance on target graphs with specific tasks, GDA introduces a bunch of task-related graphs as source graphs and adapts the knowledge learnt from source graphs to the target graphs. Since GDA combines the advantages of graph representation learning and domain adaptation, it has become a promising direction of transfer learning on graphs and has attracted an increasing amount of research interest in recent years. In this paper, we comprehensively overview the studies of GDA and present a detailed survey of recent advances. Specifically, we outline the research status and challenges, propose a taxonomy, introduce the details of representative works, and discuss the prospects. To the best of our knowledge, this paper is the first survey for graph domain adaptation. A detailed paper list is available at https://github.com/Skyorca/Awesome-Graph-Domain-Adaptation-Papers.

cs.LG↗

arXiv2023

Causality and Independence Enhancement for Biased Node Classification

Most existing methods that address out-of-distribution (OOD) generalization for node classification on graphs primarily focus on a specific type of data biases, such as label selection bias or structural bias. However, anticipating the type of bias in advance is extremely challenging, and designing models solely for one specific type may not necessarily improve overall generalization performance. Moreover, limited research has focused on the impact of mixed biases, which are more prevalent and demanding in real-world scenarios. To address these limitations, we propose a novel Causality and Independence Enhancement (CIE) framework, applicable to various graph neural networks (GNNs). Our approach estimates causal and spurious features at the node representation level and mitigates the influence of spurious correlations through the backdoor adjustment. Meanwhile, independence constraint is introduced to improve the discriminability and stability of causal and spurious features in complex biased environments. Essentially, CIE eliminates different types of data biases from a unified perspective, without the need to design separate methods for each bias as before. To evaluate the performance under specific types of data biases, mixed biases, and low-resource scenarios, we conducted comprehensive experiments on five publicly available datasets. Experimental results demonstrate that our approach CIE not only significantly enhances the performance of GNNs but outperforms state-of-the-art debiased node classification methods.

cs.LG↗

arXiv2023

Focus on the Sound around You: Monaural Target Speaker Extraction via Distance and Speaker Information

Previously, Target Speaker Extraction (TSE) has yielded outstanding performance in certain application scenarios for speech enhancement and source separation. However, obtaining auxiliary speaker-related information is still challenging in noisy environments with significant reverberation. inspired by the recently proposed distance-based sound separation, we propose the near sound (NS) extractor, which leverages distance information for TSE to reliably extract speaker information without requiring previous speaker enrolment, called speaker embedding self-enrollment (SESE). Full- & sub-band modeling is introduced to enhance our NS-Extractor's adaptability towards environments with significant reverberation. Experimental results on several cross-datasets demonstrate the effectiveness of our improvements and the excellent performance of our proposed NS-Extractor in different application scenarios.

cs.SD↗

arXiv2023

CED: Consistent ensemble distillation for audio tagging

Augmentation and knowledge distillation (KD) are well-established techniques employed in audio classification tasks, aimed at enhancing performance and reducing model sizes on the widely recognized Audioset (AS) benchmark. Although both techniques are effective individually, their combined use, called consistent teaching, hasn't been explored before. This paper proposes CED, a simple training framework that distils student models from large teacher ensembles with consistent teaching. To achieve this, CED efficiently stores logits as well as the augmentation methods on disk, making it scalable to large-scale datasets. Central to CED's efficacy is its label-free nature, meaning that only the stored logits are used for the optimization of a student model only requiring 0.3\% additional disk space for AS. The study trains various transformer-based models, including a 10M parameter model achieving a 49.0 mean average precision (mAP) on AS. Pretrained models and code are available at https://github.com/RicherMans/CED.

cs.SD↗

arXiv2023

OpenGDA: Graph Domain Adaptation Benchmark for Cross-network Learning

Graph domain adaptation models are widely adopted in cross-network learning tasks, with the aim of transferring labeling or structural knowledge. Currently, there mainly exist two limitations in evaluating graph domain adaptation models. On one side, they are primarily tested for the specific cross-network node classification task, leaving tasks at edge-level and graph-level largely under-explored. Moreover, they are primarily tested in limited scenarios, such as social networks or citation networks, lacking validation of model's capability in richer scenarios. As comprehensively assessing models could enhance model practicality in real-world applications, we propose a benchmark, known as OpenGDA. It provides abundant pre-processed and unified datasets for different types of tasks (node, edge, graph). They originate from diverse scenarios, covering web information systems, urban systems and natural systems. Furthermore, it integrates state-of-the-art models with standardized and end-to-end pipelines. Overall, OpenGDA provides a user-friendly, scalable and reproducible benchmark for evaluating graph domain adaptation models. The benchmark experiments highlight the challenges of applying GDA models to real-world applications with consistent good performance, and potentially provide insights to future research. As an emerging project, OpenGDA will be regularly updated with new datasets and models. It could be accessed from https://github.com/Skyorca/OpenGDA.

cs.AI↗

arXiv2023

AV-SepFormer: Cross-Attention SepFormer for Audio-Visual Target Speaker Extraction

Visual information can serve as an effective cue for target speaker extraction (TSE) and is vital to improving extraction performance. In this paper, we propose AV-SepFormer, a SepFormer-based attention dual-scale model that utilizes cross- and self-attention to fuse and model features from audio and visual. AV-SepFormer splits the audio feature into a number of chunks, equivalent to the length of the visual feature. Then self- and cross-attention are employed to model and fuse the multi-modal features. Furthermore, we use a novel 2D positional encoding, that introduces the positional information between and within chunks and provides significant gains over the traditional positional encoding. Our model has two key advantages: the time granularity of audio chunked feature is synchronized to the visual feature, which alleviates the harm caused by the inconsistency of audio and video sampling rate; by combining self- and cross-attention, feature fusion and speech extraction processes are unified within an attention paradigm. The experimental results show that AV-SepFormer significantly outperforms other existing methods.

cs.MM↗

arXiv2023

Understanding temporally weakly supervised training: A case study for keyword spotting

The currently most prominent algorithm to train keyword spotting (KWS) models with deep neural networks (DNNs) requires strong supervision i.e., precise knowledge of the spoken keyword location in time. Thus, most KWS approaches treat the presence of redundant data, such as noise, within their training set as an obstacle. A common training paradigm to deal with data redundancies is to use temporally weakly supervised learning, which only requires providing labels on a coarse scale. This study explores the limits of DNN training using temporally weak labeling with applications in KWS. We train a simple end-to-end classifier on the common Google Speech Commands dataset with increased difficulty by randomly appending and adding noise to the training dataset. Our results indicate that temporally weak labeling can achieve comparable results to strongly supervised baselines while having a less stringent labeling requirement. In the presence of noise, weakly supervised models are capable to localize and extract target keywords without explicit supervision, leading to a performance increase compared to strongly supervised approaches.

cs.SD↗