arXiv subjects

Ling Cheng

Publications and source records attributed to Ling Cheng.

At least 19 recordsLinked to original sources

arXiv2026

HealthLoopQA: A Context-Aware Question Answering Benchmark for Interpreting Wearable Monitoring Data in Diabetes Care

As medical wearables become integrated into daily chronic disease care, effectively interpreting longitudinal monitoring data is essential for patients and clinicians to understand health trends, detect safety-critical events, and make informed decisions. While large language models (LLMs) show promise for transforming this streaming physiological data into personalized health insights, evaluating their reasoning capability and analytical rigor in diverse monitoring tasks remains a fundamental challenge. Existing medical wearable question answering (QA) benchmarks primarily assess short-horizon classification or statistical summaries, largely ignoring the long-term patterns, therapeutic and behavioural contexts, and potential system failures inherent in real-world deployments. To address this, we introduce HealthLoopQA, a comprehensive diagnostic benchmark for evaluating LLM reasoning over continuous diabetes monitoring data. Grounded in a novel taxonomy of eleven atomic reasoning abilities, HealthLoopQA comprises 127 tasks and over 1,500 QA instances spanning process mining, anomaly detection, and prediction over 30-day horizons. To systematically evaluate safety awareness, we complement real-world datasets with a fault-injected simulation testbed modeling diverse device malfunctions and cyber-physical attacks to generate physiologically plausible hazard scenarios. Evaluating state-of-the-art LLMs across prompting and agentic frameworks reveals severe limitations in complex temporal pattern mining. Furthermore, we identify a broader phenomenon of In-context Laziness under long-context prompting, highlighting critical open challenges in deploying LLMs for rigorous long-horizon medical reasoning.

cs.LG↗

arXiv2025

LLM-based Few-Shot Early Rumor Detection with Imitation Agent

Early Rumor Detection (EARD) aims to identify the earliest point at which a claim can be accurately classified based on a sequence of social media posts. This is especially challenging in data-scarce settings. While Large Language Models (LLMs) perform well in few-shot NLP tasks, they are not well-suited for time-series data and are computationally expensive for both training and inference. In this work, we propose a novel EARD framework that combines an autonomous agent and an LLM-based detection model, where the agent acts as a reliable decision-maker for \textit{early time point determination}, while the LLM serves as a powerful \textit{rumor detector}. This approach offers the first solution for few-shot EARD, necessitating only the training of a lightweight agent and allowing the LLM to remain training-free. Extensive experiments on four real-world datasets show our approach boosts performance across LLMs and surpasses existing EARD methods in accuracy and earliness.

cs.CL↗

arXiv2025

Dynamic Electric Vehicle Charging Pricing for Load Balancing in Power Distribution Networks based on Collaborative DDPG Agents

The transition from the Internal Combustion Engine Vehicles (ICEVs) to the Electric Vehicles (EVs) is globally recommended to combat the unfavourable environmental conditions caused by reliance on fossil fuels. However, it has been established that the charging of EVs can destabilize the grid when they penetrate the market in large numbers, especially in grids that were not initially built to handle the load from the charging of EVs. In this work, we present a dynamic EV charging pricing strategy that fulfills the following three objectives: distribution network-level load peak-shaving, valley-filling, and load balancing across distribution networks. Based on historical environmental variables such as temperature, humidity, wind speed, EV charging prices and distribution of vehicles in different areas in different times of the day, we first forecast the distribution network load demand, and then use deep reinforcement learning approach to set the optimal dynamic EV charging price. While most research seeks to achieve load peak-shaving and valley-filling to stabilize the grid, our work goes further into exploring the load-balancing between the distribution networks in the close vicinity to each other. We compare the performance of Deep Deterministic Policy Gradient (DDPG), Soft Actor-Critic (SAC) and Proximal Policy Optimization (PPO) algorithms for this purpose. The best algorithm is used for dymamic EV pricing. Simulation results show an improved utilization of the grid at the distribution network level, leading to the optimal usage of the grid on a larger scale.

eess.SY↗

arXiv2025

Neural Network Model for OAM Crosstalk due to Turbulence-Induced Tilt and Lateral Displacement

Accurately modelling orbital angular momentum (OAM) mode crosstalk in turbulent environments is challenging yet essential for developing free-space optical systems that employ OAM modes for multiplexing or diversity. Turbulence induces tip/tilt aberrations and lateral displacement, which significantly degrade system performance. Existing analytical models describe the transformation from an input Gaussian mode to an output OAM spectrum; however, our feed forward neural network model generalizes this approach by accounting for the effects of these aberrations on arbitrary input OAM modes. We validate the model experimentally by estimating turbulence-induced tilt and lateral displacement using a dual-camera setup and comparing the estimated spectrum with the actual modal decomposition. With a typical root mean square error of less than 11%, our results indicate that the model could serve as a reliable source of meta-information for digital signal processing, soft-decision forward error correction or perhaps dynamic mode hopping in future systems.

physics.optics↗

arXiv2025

New Recipe for Semi-supervised Community Detection: Clique Annealing under Crystallization Kinetics

Semi-supervised community detection methods are widely used for identifying specific communities due to the label scarcity. Existing semi-supervised community detection methods typically involve two learning stages learning in both initial identification and subsequent adjustment, which often starts from an unreasonable community core candidate. Moreover, these methods encounter scalability issues because they depend on reinforcement learning and generative adversarial networks, leading to higher computational costs and restricting the selection of candidates. To address these limitations, we draw a parallel between crystallization kinetics and community detection to integrate the spontaneity of the annealing process into community detection. Specifically, we liken community detection to identifying a crystal subgrain (core) that expands into a complete grain (community) through a process similar to annealing. Based on this finding, we propose CLique ANNealing (CLANN), which applies kinetics concepts to community detection by integrating these principles into the optimization process to strengthen the consistency of the community core. Subsequently, a learning-free Transitive Annealer was employed to refine the first-stage candidates by merging neighboring cliques and repositioning the community core, enabling a spontaneous growth process that enhances scalability. Extensive experiments on \textbf{43} different network settings demonstrate that CLANN outperforms state-of-the-art methods across multiple real-world datasets, showcasing its exceptional efficacy and efficiency in community detection.

cs.SI↗

arXiv2025

Proof-of-Data: A Consensus Protocol for Collaborative Intelligence

Existing research on federated learning has been focused on the setting where learning is coordinated by a centralized entity. Yet the greatest potential of future collaborative intelligence would be unleashed in a more open and democratized setting with no central entity in a dominant role, referred to as "decentralized federated learning". New challenges arise accordingly in achieving both correct model training and fair reward allocation with collective effort among all participating nodes, especially with the threat of the Byzantine node jeopardising both tasks. In this paper, we propose a blockchain-based decentralized Byzantine fault-tolerant federated learning framework based on a novel Proof-of-Data (PoD) consensus protocol to resolve both the "trust" and "incentive" components. By decoupling model training and contribution accounting, PoD is able to enjoy not only the benefit of learning efficiency and system liveliness from asynchronous societal-scale PoW-style learning but also the finality of consensus and reward allocation from epoch-based BFT-style voting. To mitigate false reward claims by data forgery from Byzantine attacks, a privacy-aware data verification and contribution-based reward allocation mechanism is designed to complete the framework. Our evaluation results show that PoD demonstrates performance in model training close to that of the centralized counterpart while achieving trust in consensus and fairness for reward allocation with a fault tolerance ratio of 1/3.

cs.CR↗

arXiv2024

A Full-History Network Dataset for BTC Asset Decentralization Profiling

Since its advent in 2009, Bitcoin (BTC) has garnered increasing attention from both academia and industry. However, due to the massive transaction volume, no systematic study has quantitatively measured the asset decentralization degree specifically from a network perspective. In this paper, by conducting a thorough analysis of the BTC transaction network, we first address the significant gap in the availability of full-history BTC graph and network property dataset, which spans over 15 years from the genesis block (1st March, 2009) to the 845651-th block (29, May 2024). We then present the first systematic investigation to profile BTC's asset decentralization and design several decentralization degrees for quantification. Through extensive experiments, we emphasize the significant role of network properties and our network-based decentralization degree in enhancing Bitcoin analysis. Our findings demonstrate the importance of our comprehensive dataset and analysis in advancing research on Bitcoin's transaction dynamics and decentralization, providing valuable insights into the network's structure and its implications.

q-fin.ST↗

arXiv2024

Path Loss Modeling for NLoS Ultraviolet Channels Incorporating Scattering and Reflection Effects

This paper tackles limitations in existing non-line-of-sight (NLoS) ultraviolet (UV) channel models, where conventional approaches assume obstacle-free propagation or uniform radiation intensity. In this paper, we develop a path loss model incorporating scattering and reflection, and then propose an obstacle-boundary approximation method to achieve computational tractability. Our framework systematically incorporates spatial obstacle properties, including dimensions, coordinates, contours, and orientation angles, while employing the Lambertian radiation pattern for source modeling. Additionally, the proposed path loss model is validated by comparing it with the Monte-Carlo photon-tracing model and analytical integral model via numerical results, which indicate that when obstacle reflection is prominent, an approximation treatment of obstacle boundaries has a negligible influence on the path loss estimation of NLoS UV communication channels.

eess.SP↗

arXiv2024

Modeling of UV NLoS Communication Channels: From Atmospheric Scattering and Obstacle Reflection Perspectives

As transceiver elevation angles increase from small to large, existing ultraviolet (UV) non-line-of-sight (NLoS) models encounter two challenges: i) cannot estimate the channel characteristics of UV NLoS communication scenarios when there exists an obstacle in the overlap volume between the transmitter beam and the receiver field-of-view (FoV), and ii) cannot evaluate the channel path loss for the wide beam and wide FoV scenarios with existing simplified single-scattering path loss models. To address these challenges, a UV NLoS scattering model incorporating an obstacle was investigated, where the obstacle's orientation angle, coordinates, and geometric dimensions were taken into account to approach actual application environments. Then, a UV NLoS reflection model was developed combined with specific geometric diagrams. Further, a simplified single-scattering path loss model was proposed with a closed-form expression. Finally, the proposed models were validated by comparing them with the Monte-Carlo photon-tracing model, the exact single-scattering model, and the latest simplified single-scattering model. Numerical results show that the path loss curves obtained by the proposed models agree well with those attained by related NLoS models under identical parameter settings, and avoiding obstacles is not always a good option for UV NLoS communications. Moreover, the accuracy of the proposed simplified model is superior to that of the existing simplified model for all kinds of transceiver FoV angles.

eess.SP↗

arXiv2024

Channel Modeling for Ultraviolet Non-Line-of-Sight Communications Incorporating an Obstacle

Existing studies on ultraviolet (UV) non-line-of-sight (NLoS) channel modeling primarily focus on scenarios without any obstacle, which makes them unsuitable for small transceiver elevation angles in most cases. To address this issue, a UV NLoS channel model incorporating an obstacle was investigated in this paper, where the impacts of atmospheric scattering and obstacle reflection on UV signals were both taken into account. To validate the proposed model, we compared it to the related Monte-Carlo photon-tracing (MCPT) model that had been verified by outdoor experiments. Numerical results manifest that the path loss curves obtained by the proposed model agree well with those determined by the MCPT model, while its computation complexity is lower than that of the MCPT model. This work discloses that obstacle reflection can effectively reduce the channel path loss of UV NLoS communication systems.

eess.SP↗

arXiv2024

Crafting a Good Prompt or Providing Exemplary Dialogues? A Study of In-Context Learning for Persona-based Dialogue Generation

Previous in-context learning (ICL) research has focused on tasks such as classification, machine translation, text2table, etc., while studies on whether ICL can improve human-like dialogue generation are scarce. Our work fills this gap by systematically investigating the ICL capabilities of large language models (LLMs) in persona-based dialogue generation, conducting extensive experiments on high-quality real human Chinese dialogue datasets. From experimental results, we draw three conclusions: 1) adjusting prompt instructions is the most direct, effective, and economical way to improve generation quality; 2) randomly retrieving demonstrations (demos) achieves the best results, possibly due to the greater diversity and the amount of effective information; counter-intuitively, retrieving demos with a context identical to the query performs the worst; 3) even when we destroy the multi-turn associations and single-turn semantics in the demos, increasing the number of demos still improves dialogue performance, proving that LLMs can learn from corrupted dialogue demos. Previous explanations of the ICL mechanism, such as $n$-gram induction head, cannot fully account for this phenomenon.

cs.CL↗

arXiv2023

From Asset Flow to Status, Action and Intention Discovery: Early Malice Detection in Cryptocurrency

Cryptocurrency has been subject to illicit activities probably more often than traditional financial assets due to the pseudo-anonymous nature of its transacting entities. An ideal detection model is expected to achieve all three critical properties of (I) early detection, (II) good interpretability, and (III) versatility for various illicit activities. However, existing solutions cannot meet all these requirements, as most of them heavily rely on deep learning without interpretability and are only available for retrospective analysis of a specific illicit type. To tackle all these challenges, we propose Intention-Monitor for early malice detection in Bitcoin (BTC), where the on-chain record data for a certain address are much scarcer than other cryptocurrency platforms. We first define asset transfer paths with the Decision-Tree based feature Selection and Complement (DT-SC) to build different feature sets for different malice types. Then, the Status/Action Proposal Module (S/A-PM) and the Intention-VAE module generate the status, action, intent-snippet, and hidden intent-snippet embedding. With all these modules, our model is highly interpretable and can detect various illegal activities. Moreover, well-designed loss functions further enhance the prediction speed and model's interpretability. Extensive experiments on three real-world datasets demonstrate that our proposed algorithm outperforms the state-of-the-art methods. Furthermore, additional case studies justify our model can not only explain existing illicit patterns but can also find new suspicious characters.

cs.LG↗

arXiv2023

Examining the Effect of Pre-training on Time Series Classification

Although the pre-training followed by fine-tuning paradigm is used extensively in many fields, there is still some controversy surrounding the impact of pre-training on the fine-tuning process. Currently, experimental findings based on text and image data lack consensus. To delve deeper into the unsupervised pre-training followed by fine-tuning paradigm, we have extended previous research to a new modality: time series. In this study, we conducted a thorough examination of 150 classification datasets derived from the Univariate Time Series (UTS) and Multivariate Time Series (MTS) benchmarks. Our analysis reveals several key conclusions. (i) Pre-training can only help improve the optimization process for models that fit the data poorly, rather than those that fit the data well. (ii) Pre-training does not exhibit the effect of regularization when given sufficient training time. (iii) Pre-training can only speed up convergence if the model has sufficient ability to fit the data. (iv) Adding more pre-training data does not improve generalization, but it can strengthen the advantage of pre-training on the original data volume, such as faster convergence. (v) While both the pre-training task and the model structure determine the effectiveness of the paradigm on a given dataset, the model structure plays a more significant role.

cs.LG↗

arXiv2023

Code Will Tell: Visual Identification of Ponzi Schemes on Ethereum

Ethereum has become a popular blockchain with smart contracts for investors nowadays. Due to the decentralization and anonymity of Ethereum, Ponzi schemes have been easily deployed and caused significant losses to investors. However, there are still no explainable and effective methods to help investors easily identify Ponzi schemes and validate whether a smart contract is actually a Ponzi scheme. To fill the research gap, we propose PonziLens, a novel visualization approach to help investors achieve early identification of Ponzi schemes by investigating the operation codes of smart contracts. Specifically, we conduct symbolic execution of opcode and extract the control flow for investing and rewarding with critical opcode instructions. Then, an intuitive directed-graph based visualization is proposed to display the investing and rewarding flows and the crucial execution paths, enabling easy identification of Ponzi schemes on Ethereum. Two usage scenarios involving both Ponzi and non-Ponzi schemes demonstrate the effectiveness of PonziLens.

cs.HC↗

arXiv2023

Evolve Path Tracer: Early Detection of Malicious Addresses in Cryptocurrency

With the ever-increasing boom of Cryptocurrency, detecting fraudulent behaviors and associated malicious addresses draws significant research effort. However, most existing studies still rely on the full history features or full-fledged address transaction networks, thus cannot meet the requirements of early malicious address detection, which is urgent but seldom discussed by existing studies. To detect fraud behaviors of malicious addresses in the early stage, we present Evolve Path Tracer, which consists of Evolve Path Encoder LSTM, Evolve Path Graph GCN, and Hierarchical Survival Predictor. Specifically, in addition to the general address features, we propose asset transfer paths and corresponding path graphs to characterize early transaction patterns. Further, since the transaction patterns are changing rapidly during the early stage, we propose Evolve Path Encoder LSTM and Evolve Path Graph GCN to encode asset transfer path and path graph under an evolving structure setting. Hierarchical Survival Predictor then predicts addresses' labels with nice scalability and faster prediction speed. We investigate the effectiveness and versatility of Evolve Path Tracer on three real-world illicit bitcoin datasets. Our experimental results demonstrate that Evolve Path Tracer outperforms the state-of-the-art methods. Extensive scalability experiments demonstrate the model's adaptivity under a dynamic prediction setting.

cs.AI↗

arXiv2022

Toward Intention Discovery for Early Malice Detection in Bitcoin

Bitcoin has been subject to illicit activities more often than probably any other financial assets, due to the pseudo-anonymous nature of its transacting entities. An ideal detection model is expected to achieve all the three properties of (I) early detection, (II) good interpretability, and (III) versatility for various illicit activities. However, existing solutions cannot meet all these requirements, as most of them heavily rely on deep learning without satisfying interpretability and are only available for retrospective analysis of a specific illicit type. First, we present asset transfer paths, which aim to describe addresses' early characteristics. Next, with a decision tree based strategy for feature selection and segmentation, we split the entire observation period into different segments and encode each as a segment vector. After clustering all these segment vectors, we get the global status vectors, essentially the basic unit to describe the whole intention. Finally, a hierarchical self-attention predictor predicts the label for the given address in real time. A survival module tells the predictor when to stop and proposes the status sequence, namely intention. % With the type-dependent selection strategy and global status vectors, our model can be applied to detect various illicit activities with strong interpretability. The well-designed predictor and particular loss functions strengthen the model's prediction speed and interpretability one step further. Extensive experiments on three real-world datasets show that our proposed algorithm outperforms state-of-the-art methods. Besides, additional case studies justify our model can not only explain existing illicit patterns but can also find new suspicious characters.

cs.LG↗

arXiv2022

Watermark-Based Code Construction for Finite-State Markov Channel with Synchronisation Errors

With advancements in telecommunications, data transmission over increasingly harsher channels that produce synchronisation errors is inevitable. Coding schemes for such channels are available through techniques such as the Davey-MacKay watermark coding; however, this is limited to memoryless channel estimates. Memory must be accounted for to ensure a realistic channel approximation - similar to a Finite State Markov Chain or Fritchman Model. A novel code construction and decoder are developed to correct synchronisation errors while considering the channel's correlated memory effects by incorporating ideas from the watermark scheme and memory modelling. Simulation results show that the proposed code construction and decoder rival the first and second-order Davey-MacKay type watermark decoder and even perform slightly better when the inner-channel capacity is higher than 0.9. The proposed system and decoder may prove helpful in fields such as free-space optics and possibly molecular communication, where harsh channels are used for communication.

cs.IT↗

arXiv2021

Geometry-Entangled Visual Semantic Transformer for Image Captioning

Recent advancements of image captioning have featured Visual-Semantic Fusion or Geometry-Aid attention refinement. However, those fusion-based models, they are still criticized for the lack of geometry information for inter and intra attention refinement. On the other side, models based on Geometry-Aid attention still suffer from the modality gap between visual and semantic information. In this paper, we introduce a novel Geometry-Entangled Visual Semantic Transformer (GEVST) network to realize the complementary advantages of Visual-Semantic Fusion and Geometry-Aid attention refinement. Concretely, a Dense-Cap model proposes some dense captions with corresponding geometry information at first. Then, to empower GEVST with the ability to bridge the modality gap among visual and semantic information, we build four parallel transformer encoders VV(Pure Visual), VS(Semantic fused to Visual), SV(Visual fused to Semantic), SS(Pure Semantic) for final caption generation. Both visual and semantic geometry features are used in the Fusion module and also the Self-Attention module for better attention measurement. To validate our model, we conduct extensive experiments on the MS-COCO dataset, the experimental results show that our GEVST model can obtain promising performance gains.

cs.CV↗