arXiv subjects

Swee Liang Wong

Publications and source records attributed to Swee Liang Wong.

12 recordsLinked to original sources

arXiv2026

SciCodePile: A 128GB Corpus and Executable Benchmark for Challenging Scientific Code Generation

Large language models (LLMs) excel at general-purpose code generation, yet how well they handle scientific code remains an open question. Existing datasets and benchmarks are limited in scale, domain coverage, or executable verification, leaving the true gap between current LLMs and reliable scientific code generators inadequately assessed. To address these limitations, we present SciCodePile, the largest scientific code corpus to date, constructed from 37,737 public repositories and collectively comprising 128GB of code that spans multiple computational science disciplines. From this corpus, we further curate an executable benchmark of 200 tasks, each equipped with a sandboxed execution environment and an automated test harness for functional verification. We evaluate 15 LLMs from both open-source and closed-source families on three tasks: prefix-to-suffix completion, fill-in-the-middle infilling, and executable code generation. Results show that scientific code generation remains highly challenging: The best CodeBLEU reaches only 38.13 and 38.37 on the two completion tasks, while the strongest model achieves just 12.30\% Pass@1 on the executable benchmark, underscoring how far current models remain from reliable scientific code generation. To demonstrate the training utility of SciCodePile, we further show that continued pretraining on our corpus improves CodeBLEU by $\times$2.84 on scientific code completion, and instruction tuning on our data improves Pass@1 by $\times$4.79 on the executable benchmark. All code and data are available at https://huggingface.co/SciCodePile.

cs.SE↗

arXiv2026

SMT-AD: a scalable quantum-inspired anomaly detection approach

Quantum-inspired tensor networks algorithms have shown to be effective and efficient models for machine learning tasks, including anomaly detection. Here, we propose a highly parallelizable quantum-inspired approach which we call SMT-AD from Superposition of Multiresolution Tensors for Anomaly Detection. It is based upon the superposition of bond-dimension-1 matrix product operators to transform the input data with Fourier-assisted feature embedding, where the number of learnable parameters grows linearly with feature size, embedding resolutions, and the number of additional components in the matrix product operators structure. We demonstrate successful anomaly detection when applied to standard datasets, including credit card transactions, and find that, even with minimal configurations, it achieves competitive performance against established anomaly detection baselines. Furthermore, it provides a straightforward way to reduce the weight of the model and even improve the performance by highlighting the most relevant input features.

cs.LG↗

arXiv2026

Multi-Trait Subspace Steering to Reveal the Dark Side of Human-AI Interaction

Recent incidents have highlighted alarming cases where human-AI interactions led to negative psychological outcomes, including mental health crises and even user harm. As LLMs serve as sources of guidance, emotional support, and even informal therapy, these risks are poised to escalate. However, studying the mechanisms underlying harmful human-AI interactions presents significant methodological challenges, where organic harmful interactions typically develop over sustained engagement, requiring extensive conversational context that are difficult to simulate in controlled settings. To address this gap, we developed a Multi-Trait Subspace Steering (MultiTraitsss) framework that leverages established crisis-associated traits and novel subspace steering framework to generate Dark models that exhibits cumulative harmful behavioral patterns. Single-turn and multi-turn evaluations show that our dark models consistently produce harmful interaction and outcomes. Using our Dark models, we propose protective measure to reduce harmful outcomes in Human-AI interactions.

cs.AI↗

arXiv2025

Quantum Autoencoders for Anomaly Detection in Cybersecurity

Anomaly detection in cybersecurity is a challenging task, where normal events far outnumber anomalous ones with new anomalies occurring frequently. Classical autoencoders have been used for anomaly detection, but struggles in data-limited settings which quantum counterparts can potentially overcome. In this work, we apply Quantum Autoencoders (QAEs) for anomaly detection in cybersecurity, specifically on the BPF-extended tracking honeypot (BETH) dataset. QAEs are evaluated across multiple encoding techniques, ansatz types, repetitions, and feature selection strategies. Our results demonstrate that an 8-feature QAE using Dense-Angle encoding with a RealAmplitude ansatz can outperform Classical Autoencoders (CAEs), even when trained on substantially fewer samples. The effects of quantum encoding and feature selection for developing quantum models are demonstrated and discussed. In a data-limited setting, the best performing QAE model has a F1 score of 0.87, better than that of CAE (0.77). These findings suggest that QAEs may offer practical advantages for anomaly detection in data-limited scenarios.

cs.ET↗

arXiv2025

Probing Latent Subspaces in LLM for AI Security: Identifying and Manipulating Adversarial States

Large Language Models (LLMs) have demonstrated remarkable capabilities across various tasks, yet they remain vulnerable to adversarial manipulations such as jailbreaking via prompt injection attacks. These attacks bypass safety mechanisms to generate restricted or harmful content. In this study, we investigated the underlying latent subspaces of safe and jailbroken states by extracting hidden activations from a LLM. Inspired by attractor dynamics in neuroscience, we hypothesized that LLM activations settle into semi stable states that can be identified and perturbed to induce state transitions. Using dimensionality reduction techniques, we projected activations from safe and jailbroken responses to reveal latent subspaces in lower dimensional spaces. We then derived a perturbation vector that when applied to safe representations, shifted the model towards a jailbreak state. Our results demonstrate that this causal intervention results in statistically significant jailbreak responses in a subset of prompts. Next, we probed how these perturbations propagate through the model's layers, testing whether the induced state change remains localized or cascades throughout the network. Our findings indicate that targeted perturbations induced distinct shifts in activations and model responses. Our approach paves the way for potential proactive defenses, shifting from traditional guardrail based methods to preemptive, model agnostic techniques that neutralize adversarial states at the representation level.

cs.LG↗

arXiv2025

Prompt Inject Detection with Generative Explanation as an Investigative Tool

Large Language Models (LLMs) are vulnerable to adversarial prompt based injects. These injects could jailbreak or exploit vulnerabilities within these models with explicit prompt requests leading to undesired responses. In the context of investigating prompt injects, the challenge is the sheer volume of input prompts involved that are likely to be largely benign. This investigative challenge is further complicated by the semantics and subjectivity of the input prompts involved in the LLM conversation with its user and the context of the environment to which the conversation is being carried out. Hence, the challenge for AI security investigators would be two-fold. The first is to identify adversarial prompt injects and then to assess whether the input prompt is contextually benign or adversarial. For the first step, this could be done using existing AI security solutions like guardrails to detect and protect the LLMs. Guardrails have been developed using a variety of approaches. A popular approach is to use signature based. Another popular approach to develop AI models to classify such prompts include the use of NLP based models like a language model. However, in the context of conducting an AI security investigation of prompt injects, these guardrails lack the ability to aid investigators in triaging or assessing the identified input prompts. In this applied research exploration, we explore the use of a text generation capabilities of LLM to detect prompt injects and generate explanation for its detections to aid AI security investigators in assessing and triaging of such prompt inject detections. The practical benefit of such a tool is to ease the task of conducting investigation into prompt injects.

cs.CR↗

arXiv2024

Audio Simulation for Sound Source Localization in Virtual Evironment

Non-line-of-sight localization in signal-deprived environments is a challenging yet pertinent problem. Acoustic methods in such predominantly indoor scenarios encounter difficulty due to the reverberant nature. In this study, we aim to locate sound sources to specific locations within a virtual environment by leveraging physically grounded sound propagation simulations and machine learning methods. This process attempts to overcome the issue of data insufficiency to localize sound sources to their location of occurrence especially in post-event localization. We achieve 0.786+/- 0.0136 F1-score using an audio transformer spectrogram approach.

cs.LG↗

arXiv2024

Enhancing Reasoning Capacity of SLM using Cognitive Enhancement

Large Language Models (LLMs) have been applied to automate cyber security activities and processes including cyber investigation and digital forensics. However, the use of such models for cyber investigation and digital forensics should address accountability and security considerations. Accountability ensures models have the means to provide explainable reasonings and outcomes. This information can be extracted through explicit prompt requests. For security considerations, it is crucial to address privacy and confidentiality of the involved data during data processing as well. One approach to deal with this consideration is to have the data processed locally using a local instance of the model. Due to limitations of locally available resources, namely memory and GPU capacities, a Smaller Large Language Model (SLM) will typically be used. These SLMs have significantly fewer parameters compared to the LLMs. However, such size reductions have notable performance reduction, especially when tasked to provide reasoning explanations. In this paper, we aim to mitigate performance reduction through the integration of cognitive strategies that humans use for problem-solving. We term this as cognitive enhancement through prompts. Our experiments showed significant improvement gains of the SLMs' performances when such enhancements were applied. We believe that our exploration study paves the way for further investigation into the use of cognitive enhancement to optimize SLM for cyber security applications.

cs.CR↗

arXiv2023

RAGLog: Log Anomaly Detection using Retrieval Augmented Generation

The ability to detect log anomalies from system logs is a vital activity needed to ensure cyber resiliency of systems. It is applied for fault identification or facilitate cyber investigation and digital forensics. However, as logs belonging to different systems and components differ significantly, the challenge to perform such analysis is humanly challenging from the volume, variety and velocity of logs. This is further complicated by the lack or unavailability of anomalous log entries to develop trained machine learning or artificial intelligence models for such purposes. In this research work, we explore the use of a Retrieval Augmented Large Language Model that leverages a vector database to detect anomalies from logs. We used a Question and Answer configuration pipeline. To the best of our knowledge, our experiment which we called RAGLog is a novel one and the experimental results show much promise.

cs.CR↗

arXiv2020

Machine learning and high-throughput robust design of P3HT-CNT composite thin films for high electrical conductivity

Combining high-throughput experiments with machine learning allows quick optimization of parameter spaces towards achieving target properties. In this study, we demonstrate that machine learning, combined with multi-labeled datasets, can additionally be used for scientific understanding and hypothesis testing. We introduce an automated flow system with high-throughput drop-casting for thin film preparation, followed by fast characterization of optical and electrical properties, with the capability to complete one cycle of learning of fully labeled ~160 samples in a single day. We combine regio-regular poly-3-hexylthiophene with various carbon nanotubes to achieve electrical conductivities as high as 1200 S/cm. Interestingly, a non-intuitive local optimum emerges when 10% of double-walled carbon nanotubes are added with long single wall carbon nanotubes, where the conductivity is seen to be as high as 700 S/cm, which we subsequently explain with high fidelity optical characterization. Employing dataset resampling strategies and graph-based regressions allows us to account for experimental cost and uncertainty estimation of correlated multi-outputs, and supports the proving of the hypothesis linking charge delocalization to electrical conductivity. We therefore present a robust machine-learning driven high-throughput experimental scheme that can be applied to optimize and understand properties of composites, or hybrid organic-inorganic materials.

physics.app-ph↗

arXiv2019

Coulomb blockade in Etched Single and Few Layer MoS2 Nanoribbons

Confinement in two-dimensional transition metal dichalcogenides is an attractive platform for trapping single charge and spins for quantum information processing. Here, we present low temperature electron transport through etched 50-70nm MoS2 nanoribbons showing current oscillations as a function of gate voltage. On further investigations current through the device forms diamond shaped domains as a function of source-drain and gate voltage. We associate these current oscillations and diamond shaped current domains with Coulomb blockade due to single electron tunneling through a quantum dot formed in the MoS2 nanoribbon. From the size of the Coulomb diamond, we estimate the quantum dot size as small as 10-35nm. We discuss the possible origins of quantum dot in our nanoribbon device and prospects to control or engineer the quantum dot in such etched MoS2 nanoribbons which can be a promising platform for spin-valley qubits in two-dimensional transition metal dichalcogenides.

cond-mat.mes-hall↗

arXiv2019

Momentum Conserved Ultrafast Charge Transfer Dynamics of Interlayer Excitons in vdW Heterostructures

Heterostructures comprising van der Waals (vdW) stacked transition metal dichalcogenide (TMDC) monolayers are a fascinating class of two-dimensional (2D) materials with unique properties. The presence of interlayer excitons, where the electron and the hole remain spatially separated in the two layers due to ultrafast charge transfer, is an intriguing feature of these heterostructures. Inevitably, the efficiency of 2D heterostructure devices is critically dependent on the charge transfer dynamics. However, the role of the relative rotation angle of the constituent layers on this charge transfer dynamics is hitherto unknown. Here, we investigate MoS$_2$/WSe$_2$ vdW heterostructures (hMWs) using monochromated low-loss electron energy loss (EEL) spectroscopy combined with aberration-corrected scanning transmission electron microscopy (STEM), and report that momentum conservation is a critical factor in the charge transfer dynamics of TMDC vdW heterostructures. The low-loss EEL spectra of the heterostructures with various rotation angles reveal that the charge transfer rate can be about one order-of-magnitude faster in the aligned (or anti-aligned) case than the misaligned cases. These results provide a deeper insight into the role of the fundamental principle of momentum conservation in 2D vdW heterostructure charge transfer dynamics.

cond-mat.mes-hall↗