arXiv subjects

Henry Leung

Publications and source records attributed to Henry Leung.

At least 19 recordsLinked to original sources

arXiv2026

Structural Rationale Distillation via Reasoning Space Compression

When distilling reasoning from large language models (LLMs) into smaller ones, teacher rationales for similar problems often vary wildly in structure and strategy. Like a chef who makes the same dish differently each time, this inconsistency burdens the student with noisy supervision that is hard to internalize. We propose Distillation through Reasoning Path Compression (D-RPC), which constrains the teacher to follow a compact, dynamically maintained bank of reusable high-level reasoning paths. For each training question, D-RPC retrieves the most relevant path and conditions the teacher to follow it, producing rationales that are consistent across similar problems yet diverse enough to cover different problem types. A PAC-Bayes analysis formalizes the resulting trade-off between bank size and coverage: smaller banks reduce supervision entropy but risk coverage gaps, and the generalization bound identifies an optimal intermediate size confirmed by our ablations. Across five math and commonsense reasoning benchmarks with two student models, D-RPC consistently outperforms chain-of-thought distillation, freeform rationale generation, direct distillation, and structured-supervision baselines, while using fewer tokens than template-heavy alternatives.

cs.CL↗

arXiv2025

Bias Testing and Mitigation in Black Box LLMs using Metamorphic Relations

The widespread deployment of Large Language Models (LLMs) has intensified concerns about subtle social biases embedded in their outputs. Existing guardrails often fail when faced with indirect or contextually complex bias-inducing prompts. To address these limitations, we propose a unified framework for both systematic bias evaluation and targeted mitigation. Our approach introduces six novel Metamorphic Relations (MRs) that, based on metamorphic testing principles, transform direct bias-inducing inputs into semantically equivalent yet adversarially challenging variants. These transformations enable an automated method for exposing hidden model biases: when an LLM responds inconsistently or unfairly across MR-generated variants, the underlying bias becomes detectable. We further show that the same MRs can be used to generate diverse bias-inducing samples for fine-tuning, directly linking the testing process to mitigation. Using six state-of-the-art LLMs - spanning open-source and proprietary models - and a representative subset of 385 questions from the 8,978-item BiasAsker benchmark covering seven protected groups, our MRs reveal up to 14% more hidden biases compared to existing tools. Moreover, fine-tuning with both original and MR-mutated samples significantly enhances bias resiliency, increasing safe response rates from 54.7% to over 88.9% across models. These results highlight metamorphic relations as a practical mechanism for improving fairness in conversational AI.

cs.SE↗

arXiv2025

Tackling the Kidnapped Robot Problem via Sparse Feasible Hypothesis Sampling and Reliable Batched Multi-Stage Inference

This paper addresses the Kidnapped Robot Problem (KRP), a core localization challenge of relocalizing a robot in a known map without prior pose estimate upon localization loss or at SLAM initialization. For this purpose, a passive 2-D global relocalization framework is proposed. It estimates the global pose efficiently and reliably from a single LiDAR scan and an occupancy grid map while the robot remains stationary, thereby enhancing the long-term autonomy of mobile robots. The proposed framework casts global relocalization as a non-convex problem and solves it via the multi-hypothesis scheme with batched multi-stage inference and early termination, balancing completeness and efficiency. The Rapidly-exploring Random Tree (RRT), under traversability constraints, asymptotically covers the reachable space to generate sparse, uniformly distributed feasible positional hypotheses, fundamentally reducing the sampling space. The hypotheses are preliminarily ordered by the proposed Scan Mean Absolute Difference (SMAD), a coarse beam-error level metric that facilitates the early termination by prioritizing high-likelihood candidates. The SMAD computation is optimized for limited scan measurements. The Translation-Affinity Scan-to-Map Alignment Metric (TAM) is proposed for reliable orientation selection at hypothesized positions and accurate final global pose evaluation to mitigate degradation in conventional likelihood-field metrics under translational uncertainty induced by sparse hypotheses, as well as non-panoramic LiDAR scan and environmental changes. Real-world experiments on a resource-constrained mobile robot with non-panoramic LiDAR scans show that the proposed framework achieves competitive performance in success rate, robustness under measurement uncertainty, and computational efficiency.

cs.RO↗

arXiv2025

Dual-stage and Lightweight Patient Chart Summarization for Emergency Physicians

Electronic health records (EHRs) contain extensive unstructured clinical data that can overwhelm emergency physicians trying to identify critical information. We present a two-stage summarization system that runs entirely on embedded devices, enabling offline clinical summarization while preserving patient privacy. In our approach, a dual-device architecture first retrieves relevant patient record sections using the Jetson Nano-R (Retrieve), then generates a structured summary on another Jetson Nano-S (Summarize), communicating via a lightweight socket link. The summarization output is two-fold: (1) a fixed-format list of critical findings, and (2) a context-specific narrative focused on the clinician's query. The retrieval stage uses locally stored EHRs, splits long notes into semantically coherent sections, and searches for the most relevant sections per query. The generation stage uses a locally hosted small language model (SLM) to produce the summary from the retrieved text, operating within the constraints of two NVIDIA Jetson devices. We first benchmarked six open-source SLMs under 7B parameters to identify viable models. We incorporated an LLM-as-Judge evaluation mechanism to assess summary quality in terms of factual accuracy, completeness, and clarity. Preliminary results on MIMIC-IV and de-identified real EHRs demonstrate that our fully offline system can effectively produce useful summaries in under 30 seconds.

cs.CL↗

arXiv2025

SPEAR: Soft Prompt Enhanced Anomaly Recognition for Time Series Data

Time series anomaly detection plays a crucial role in a wide range of fields, such as healthcare and internet traffic monitoring. The emergence of large language models (LLMs) offers new opportunities for detecting anomalies in the ubiquitous time series data. Traditional approaches struggle with variable-length time series sequences and context-based anomalies. We propose Soft Prompt Enhanced Anomaly Recognition (SPEAR), a novel approach to leverage LLMs for anomaly detection with soft prompts and quantization. Our methodology involves quantizing and transforming the time series data into input embeddings and combining them with learnable soft prompt embeddings. These combined embeddings are then fed into a frozen LLM. The soft prompts are updated iteratively based on a cross-entropy loss, allowing the model to adapt to time series anomaly detection. The use of soft prompts helps adapt LLMs effectively to time series tasks, while quantization ensures optimal handling of sequences, as LLMs are designed to handle discrete sequences. Our experimental results demonstrate that soft prompts effectively increase LLMs' performance in downstream tasks regarding time series anomaly detection.

cs.LG↗

arXiv2025

Towards Carbon-Aware Container Orchestration: Predicting Workload Energy Consumption with Federated Learning

The growing reliance on large-scale data centers to run resource-intensive workloads has significantly increased the global carbon footprint, underscoring the need for sustainable computing solutions. While container orchestration platforms like Kubernetes help optimize workload scheduling to reduce carbon emissions, existing methods often depend on centralized machine learning models that raise privacy concerns and struggle to generalize across diverse environments. In this paper, we propose a federated learning approach for energy consumption prediction that preserves data privacy by keeping sensitive operational data within individual enterprises. By extending the Kubernetes Efficient Power Level Exporter (Kepler), our framework trains XGBoost models collaboratively across distributed clients using Flower's FedXgbBagging aggregation using a bagging strategy, eliminating the need for centralized data sharing. Experimental results on the SPECPower benchmark dataset show that our FL-based approach achieves 11.7 percent lower Mean Absolute Error compared to a centralized baseline. This work addresses the unresolved trade-off between data privacy and energy prediction efficiency in prior systems such as Kepler and CASPER and offers enterprises a viable pathway toward sustainable cloud computing without compromising operational privacy.

cs.DC↗

arXiv2025

Bitcoin Cross-Chain Bridge: A Taxonomy and Its Promise in Artificial Intelligence of Things

Bitcoin's limited scripting capabilities and lack of native interoperability mechanisms have constrained its integration into the broader blockchain ecosystem, especially decentralized finance (DeFi) and multi-chain applications. This paper presents a comprehensive taxonomy of Bitcoin cross-chain bridge protocols, systematically analyzing their trust assumptions, performance characteristics, and applicability to the Artificial Intelligence of Things (AIoT) scenarios. We categorize bridge designs into three main types: naive token swapping, pegged-asset bridges, and arbitrary-message bridges. Each category is evaluated across key metrics such as trust model, latency, capital efficiency, and DeFi composability. Emerging innovations like BitVM and recursive sidechains are highlighted for their potential to enable secure, scalable, and programmable Bitcoin interoperability. Furthermore, we explore practical use cases of cross-chain bridges in AIoT applications, including decentralized energy trading, healthcare data integration, and supply chain automation. This taxonomy provides a foundational framework for researchers and practitioners seeking to design secure and efficient cross-chain infrastructures in AIoT systems.

cs.CR↗

arXiv2025

A Survey on Task Scheduling in Carbon-Aware Container Orchestration

The soaring energy demands of large-scale software ecosystems and cloud data centers, accelerated by the intensive training and deployment of large language models, have driven energy consumption and carbon footprint to unprecedented levels. In response, both industry and academia are increasing efforts to reduce the carbon emissions associated with cloud computing through more efficient task scheduling and infrastructure orchestration. In this work, we present a systematic review of various Kubernetes scheduling strategies, categorizing them into hardware-centric and software-centric, annotating each with its sustainability objectives, and grouping them according to the algorithms they use. We propose a comprehensive taxonomy for cloud task scheduling studies, with a particular focus on the environmental sustainability aspect. We analyze emerging research trends and open challenges, and our findings provide critical insight into the design of sustainable scheduling solutions for next-generation cloud computing systems.

cs.SE↗

arXiv2025

Globalization for Scalable Short-term Load Forecasting

Forecasting load in power transmission networks is essential across various hierarchical levels, from the system level down to individual points of delivery (PoD). While intuitive and locally accurate, traditional local forecasting models (LFMs) face significant limitations, particularly in handling generalizability, overfitting, data drift, and the cold start problem. These methods also struggle with scalability, becoming computationally expensive and less efficient as the network's size and data volume grow. In contrast, global forecasting models (GFMs) offer a new approach to enhance prediction generalizability, scalability, accuracy, and robustness through globalization and cross-learning. This paper investigates global load forecasting in the presence of data drifts, highlighting the impact of different modeling techniques and data heterogeneity. We explore feature-transforming and target-transforming models, demonstrating how globalization, data heterogeneity, and data drift affect each differently. In addition, we examine the role of globalization in peak load forecasting and its potential for hierarchical forecasting. To address data heterogeneity and the balance between globality and locality, we propose separate time series clustering (TSC) methods, introducing model-based TSC for feature-transforming models and new weighted instance-based TSC for target-transforming models. Through extensive experiments on a real-world dataset of Alberta's electricity load, we demonstrate that global target-transforming models consistently outperform their local counterparts, especially when enriched with global features and clustering techniques. In contrast, global feature-transforming models face challenges in balancing local and global dynamics, often requiring TSC to manage data heterogeneity effectively.

cs.LG↗

arXiv2025

Autonomous Exploration-Based Precise Mapping for Mobile Robots through Stepwise and Consistent Motions

This paper presents an autonomous exploration framework. It is designed for indoor ground mobile robots that utilize laser Simultaneous Localization and Mapping (SLAM), ensuring process completeness and precise mapping results. For frontier search, the local-global sampling architecture based on multiple Rapidly Exploring Random Trees (RRTs) is employed. Traversability checks during RRT expansion and global RRT pruning upon map updates eliminate unreachable frontiers, reducing potential collisions and deadlocks. Adaptive sampling density adjustments, informed by obstacle distribution, enhance exploration coverage potential. For frontier point navigation, a stepwise consistent motion strategy is adopted, wherein the robot strictly drives straight on approximately equidistant line segments in the polyline path and rotates in place at segment junctions. This simplified, decoupled motion pattern improves scan-matching stability and mitigates map drift. For process control, the framework serializes frontier point selection and navigation, avoiding oscillation caused by frequent goal changes in conventional parallelized processes. The waypoint retracing mechanism is introduced to generate repeated observations, triggering loop closure detection and backend optimization in graph-based SLAM, thereby improving map consistency and precision. Experiments in both simulation and real-world scenarios validate the effectiveness of the framework. It achieves improved mapping coverage and precision in more challenging environments compared to baseline 2D exploration algorithms. It also shows robustness in supporting resource-constrained robot platforms and maintaining mapping consistency across various LiDAR field-of-view (FoV) configurations.

cs.RO↗

arXiv2025

PCS: Perceived Confidence Scoring of Black Box LLMs with Metamorphic Relations

Zero-shot LLMs are now also used for textual classification tasks, e.g., sentiment and bias detection in a sentence or article. However, their performance can be suboptimal in such data annotation tasks. We introduce a novel technique that evaluates an LLM's confidence for classifying a textual input by leveraging Metamorphic Relations (MRs). The MRs generate semantically equivalent yet textually divergent versions of the input. Following the principles of Metamorphic Testing (MT), the mutated versions are expected to have annotation labels similar to the input. By analyzing the consistency of an LLM's responses across these variations, we compute a perceived confidence score (PCS) based on the frequency of the predicted labels. PCS can be used for both single and multiple LLM settings (e.g., when multiple LLMs are vetted in a majority-voting setup). Empirical evaluation shows that our PCS-based approach improves the performance of zero-shot LLMs by 9.3% in textual classification tasks. When multiple LLMs are used in a majority-voting setup, we obtain a performance boost of 5.8% with PCS.

cs.CL↗

arXiv2024

Unifying Tree-Reweighted Belief Propagation and Mean Field for Tracking Extended Targets

This paper proposes a unified tree-reweighted belief propagation (BP) and mean field (MF) approach for scalable detection and tracking of extended targets within the framework of factor graph. The factor graph is partitioned into a BP region and an MF region so that the messages in each region are updated according to the corresponding region rules. The BP region exploits the tree-reweighted BP, which offers improved convergence than the standard BP for graphs with massive cycles, to resolve data association. The MF region approximates the posterior densities of the measurement rate, kinematic state and extent. For linear Gaussian target models and gamma Gaussian inverse Wishart distributed state density, the unified approach provides a closed-form recursion for the state density. Hence, the proposed algorithm is more efficient than particle-based BP algorithms for extended target tracking. This method also avoids measurement clustering and gating since it solves the data association problem in a probabilistic fashion. We compare the proposed approach with algorithms such as the Poisson multi-Bernoulli mixture filter and the BP-based Poisson multi-Bernoulli filter. Simulation results demonstrate that the proposed algorithm achieves enhanced tracking performance.

eess.SP↗

arXiv2024

New bulk cone singularities in Vaidya-like spacetimes from large $c$ conformal blocks

Bulk cone singularities are singularities in boundary two-point functions at points separated by a null geodesic in the bulk, but not in the boundary. In this work, we describe a new type of bulk cone singularities in a family of Vaidya-like spacetimes that are labeled by the radius $r_+$ of the resulting black hole. We find a sharp transition in the causal structure within this family of spacetimes at $r_+=l$, the AdS length. In particular, there are bulk cone singularities that do not exist in the $r_+>l$ case, but appear for $r_+<l$. In the case of $r_+<l$ where such singularities exist, we are able to reproduce the singularities in a CFT$_2$ calculation using large $c$ conformal blocks.

hep-th↗

arXiv2024

Navigating High-Degree Heterogeneity: Federated Learning in Aerial and Space Networks

Federated learning offers a compelling solution to the challenges of networking and data privacy within aerial and space networks by utilizing vast private edge data and computing capabilities accessible through drones, balloons, and satellites. While current research has focused on optimizing the learning process, computing efficiency, and minimizing communication overhead, the heterogeneity issue and class imbalance remain a significant barrier to rapid model convergence. In this paper, we explore the influence of heterogeneity on class imbalance, which diminishes performance in Aerial and Space Networks (ASNs)-based federated learning. We illustrate the correlation between heterogeneity and class imbalance within grouped data and show how constraints such as battery life exacerbate the class imbalance challenge. Our findings indicate that ASNs-based FL faces heightened class imbalance issues even with similar levels of heterogeneity compared to other scenarios. Finally, we analyze the impact of varying degrees of heterogeneity on FL training and evaluate the efficacy of current state-of-the-art algorithms under these conditions. Our results reveal that the heterogeneity challenge is more pronounced in ASNs-based federated learning and that prevailing algorithms often fail to effectively address high levels of heterogeneity.

cs.LG↗

arXiv2024

LoRaWAN Based Dynamic Noise Mapping with Machine Learning for Urban Noise Enforcement

Static noise maps depicting long-term noise levels over wide areas are valuable urban planning assets for municipalities in decreasing noise exposure of residents. However, non-traffic noise sources with transient behavior, which people complain frequently, are usually ignored by static maps. We propose here a dynamic noise mapping approach using the data collected via low-power wide-area network (LPWAN, specifically LoRaWAN) based internet of things (IoT) infrastructure, which is one of the most common communication backbones for smart cities. Noise mapping based on LPWAN is challenging due to the low data rates of these protocols. The proposed dynamic noise mapping approach diminishes the negative implications of data rate limitations using machine learning (ML) for event and location prediction of non-traffic sources based on the scarce data. The strength of these models lies in their consideration of the spatial variance in acoustic behavior caused by the buildings in urban settings. The effectiveness of the proposed method and the accuracy of the resulting dynamic maps are evaluated in field tests. The results show that the proposed system can decrease the map error caused by non-traffic sources up to 51% and can stay effective under significant packet losses.

cs.AI↗

arXiv2024

FedGreen: Carbon-aware Federated Learning with Model Size Adaptation

Federated learning (FL) provides a promising collaborative framework to build a model from distributed clients, and this work investigates the carbon emission of the FL process. Cloud and edge servers hosting FL clients may exhibit diverse carbon footprints influenced by their geographical locations with varying power sources, offering opportunities to reduce carbon emissions by training local models with adaptive computations and communications. In this paper, we propose FedGreen, a carbon-aware FL approach to efficiently train models by adopting adaptive model sizes shared with clients based on their carbon profiles and locations using ordered dropout as a model compression technique. We theoretically analyze the trade-offs between the produced carbon emissions and the convergence accuracy, considering the carbon intensity discrepancy across countries to choose the parameters optimally. Empirical studies show that FedGreen can substantially reduce the carbon footprints of FL compared to the state-of-the-art while maintaining competitive model accuracy.

cs.LG↗

arXiv2024

Federated Learning Model Aggregation in Heterogenous Aerial and Space Networks

Federated learning offers a promising approach under the constraints of networking and data privacy constraints in aerial and space networks (ASNs), utilizing large-scale private edge data from drones, balloons, and satellites. Existing research has extensively studied the optimization of the learning process, computing efficiency, and communication overhead. An important yet often overlooked aspect is that participants contribute predictive knowledge with varying diversity of knowledge, affecting the quality of the learned federated models. In this paper, we propose a novel approach to address this issue by introducing a Weighted Averaging and Client Selection (WeiAvgCS) framework that emphasizes updates from high-diversity clients and diminishes the influence of those from low-diversity clients. Direct sharing of the data distribution may be prohibitive due to the additional private information that is sent from the clients. As such, we introduce an estimation for the diversity using a projection-based method. Extensive experiments have been performed to show WeiAvgCS's effectiveness. WeiAvgCS could converge 46% faster on FashionMNIST and 38% faster on CIFAR10 than its benchmarks on average in our experiments.

cs.LG↗

arXiv2024

Boundary signature of singularity in the presence of a shock wave

Matter falling into a Schwarzschild-AdS black hole from the left causes increased focussing of ingoing geodesics from the right, and, as a consequence, they reach the singularity sooner. In a standard Penrose diagram, the singularity "bends down". We show how to detect this feature of the singularity holographically, using a boundary two-point function. We model the matter with a shock wave, and show that this bending down of the singularity can be read off from a novel analytic continuation of the boundary two-point function. Along the way, we obtain a generalization of the recently proposed thermal product formula for two-point correlators.

hep-th↗