arXiv subjects

David Williams

Publications and source records attributed to David Williams.

At least 19 recordsLinked to original sources

arXiv2026

A Multiwavelength Study of a Long-Duration VHE Flare from BL Lacertae with VERITAS

We report the first observations of a long-duration very-high-energy (VHE; $E > 100$ GeV) flare from BL Lacertae (VER J2202+422), taken with the Very Energetic Radiation Imaging Telescope Array System (VERITAS). On October 15, 2022, the Fermi-Large Area Telescope (LAT) detected elevated GeV activity originating from this blazar. This triggered a multiwavelength campaign, which includes observations from VERITAS, Swift, NuSTAR, and select optical and radio observatories. VERITAS observed the source for a total of $\sim 9.8$ hours between September 1, 2022 and December 1, 2022. An analysis of these data yields a $\sim 28 \sigma$ detection of the source. While previously observed VHE flares from BL Lacertae have lasted on time-scales of minutes to days, VERITAS continued to detect flaring activity from the source for over a month ($\sim 40$ days) after the original flaring activity was detected with Fermi-LAT. Broadband spectral modeling shows that a synchrotron self-Compton (SSC) model with an external inverse-Compton (EC) component is preferred over a one-zone SSC model.

astro-ph.HE↗

arXiv2026

How Do Practitioners Build SE Agents? Insights from a Mixed-Methods Study

The rise of Software Engineering (SE) agents, i.e., LLM-based agents that can understand large codebases and carry out engineering tasks with limited human intervention, has been marked by rapid advances and adoption, but little is known about how developers build these systems in practice: existing studies mine repositories or examine deployment, but few investigate how SE agents are constructed. Through semi-structured interviews with 20 practitioners from 12 organizations and an online survey of 80 practitioners, this paper is the first to study how SE processes are changing in the development of SE agents and what challenges developers face. We find that as implementation becomes cheaper, bottlenecks shift rather than disappear: long-standing work in requirements, coordination, and deployment becomes more visible, while reviewing generated code and evaluating agent behavior become new and increasingly central forms of work. We characterize a seven-stage workflow and five process shifts, including a move toward evaluation-driven development, in which evaluation is increasingly defined early and steers iteration, and the emergence of specifications as first-class artifacts that teams test and version alongside code. We further identify six challenges that teams face, together with 12 corresponding practices they use or propose to address them, including unreliable evaluation signals, comprehension debt as code outpaces understanding, and behavioral changes introduced by provider-side model updates.

cs.SE↗

arXiv2026

Pomona: Continuous Code Quality Improvement via Small, Agentic Pull Requests at Bloomberg

In this industrial experience paper, we present Pomona, a lightweight agentic tool that utilises agent skills for continuous code quality improvement. Inspired by the Kaizen (TM) philosophy, Pomona automates a cycle of discovery and incremental repair: a Scanning skill identifies tasks and prioritises them in a backlog, while a Repair skill generates small, easily reviewable pull requests (PRs). This design enables frequent, low-risk improvements while maintaining engineer trust and reducing technical debt. We evaluated Pomona at Bloomberg through a three-month team deployment and a questionnaire distributed to senior engineers. The results are promising: 32/39 PRs (82.1%) were merged with a median time-to-close of just over two hours. Moreover, 10/12 surveyed engineers expressed a desire to adopt Pomona, praising its small diff sizes and focus on code quality. Since our evaluation, another team has adopted Pomona. We conclude with actionable insights for deploying agents in industry.

cs.SE↗

arXiv2026

Binomial Edge Ideals of K\"onig Type

We first characterise graphs with binomial edge ideals of K\"onig type as those for which the path covering number is equal to a minor variant of the scattering number. This enables us to apply known graph-theoretic results to immediately deduce that several classes of graphs have binomial edge ideals of K\"onig type. In particular, we show this for cocomparability graphs, or weakly closed graphs in the language of Matsuda. Along with work of LaClair and McCullough, this allows us to prove that an unmixed binomial edge ideal is of K\"onig type if and only if G is weakly closed. We then conjecture that AT-free graphs have binomial edge ideals of K\"onig type.

math.AC↗

arXiv2026

SafeTune: Search-based Harmfulness Minimisation for Large Language Models

The widespread adoption of Large Language Models (LLMs) raises concerns about the potential harmfulness of their responses. In this paper, we first investigate the harmfulness of responses from four general-purpose LLMs. Next, we propose SafeTune, a multi-objective search-based approach to mitigate harmfulness while increasing response relevance through hyperparameter tuning and system prompt engineering. Our initial evaluation shows that SafeTune significantly reduces the rate of harmful responses generated by Qwen3.5 0.8B and increases prompt-response relevance (both with a large effect size). Among the parameters we explore, we also find that encouraging greater repetition in responses is most impactful in reducing harmfulness while increasing relevance.

cs.SE↗

arXiv2026

BayesInsights: Modelling Software Delivery and Developer Experience with Bayesian Networks at Bloomberg

As software in industry grows in size and complexity, so does the volume of engineering data that companies generate and use. Ideally, this data could be used for many purposes, including informing decisions on engineering priorities. However, without a structured representation of the links between different aspects of software development, companies can struggle to identify the root causes of deficiencies or anticipate the effects of changes. In this paper, we report on our experience at Bloomberg in developing a novel tool, dubbed BayesInsights, which provides an interactive interface for visualising causal dependencies across various aspects of the software engineering (SE) process using Bayesian Networks (BNs). We describe our journey from defining network structures using a combination of established literature, expert insight, and structure learning algorithms, to integrating BayesInsights into existing data analytics solutions, and conclude with a mixed-methods evaluation of performance benchmarking and survey responses from 24 senior practitioners at Bloomberg. Our results revealed 95.8% of participants found the tool useful for identifying software delivery challenges at the team and organisational levels, cementing its value as a proof of concept for modelling software delivery and developer experience. BayesInsights is currently in preview, with access granted to seven engineering teams and a wider deployment roadmap in place for the future.

cs.SE↗

arXiv2026

Unveiling Practical Shortcomings of Patch Overfitting Detection Techniques

Automated Program Repair (APR) can reduce the time developers spend debugging, allowing them to focus on other aspects of software development. Automatically generated bug patches are typically validated through software testing. However, this method can lead to patch overfitting, i.e., generating patches that pass the given tests but are still incorrect. Patch correctness assessment (also known as overfitting detection) techniques have been proposed to identify patches that overfit. However, prior work often assessed the effectiveness of these techniques in isolation and on datasets that do not reflect the distribution of correct-to-overfitting patches that would be generated by APR tools in typical use; thus, we still do not know their effectiveness in practice. This work presents the first comprehensive benchmarking study of several patch overfitting detection (POD) methods in a practical scenario. To this end, we curate datasets that reflect realistic assumptions (i.e., patches produced by tools run under the same experimental conditions). Next, we use these data to benchmark six state-of-the-art POD approaches -- spanning static analysis, dynamic testing, and learning-based approaches -- against two baselines based on random sampling (one from prior work and one proposed herein). Our results are striking: Simple random selection outperforms all POD tools for 71% to 96% of cases, depending on the POD tool. This suggests two main takeaways: (1) current POD tools offer limited practical benefit, highlighting the need for novel techniques; (2) any POD tool must be benchmarked on realistic data and against random sampling to prove its practical effectiveness. To this end, we encourage the APR community to continue improving POD techniques and to adopt our proposed methodology for practical benchmarking; we make our data and code available to facilitate such adoption.

cs.SE↗

arXiv2026

Comparing AI Coding Agents: A Task-Stratified Analysis of Pull Request Acceptance

The rapid adoption of AI-powered coding assistants is transforming software development practices, yet systematic comparisons of their effectiveness across different task types and over time remain limited. This paper presents an empirical study comparing five popular agents (OpenAI Codex, GitHub Copilot, Devin, Cursor, and Claude Code), analyzing 7,156 pull requests (PRs) from the AIDev dataset. Temporal trend analysis reveals heterogeneous evolution patterns: Devin exhibits the only consistent positive trend in acceptance rate (+0.77% per week over 32 weeks), whereas other agents remain largely stable. Our analysis suggests that the PR task type is a dominant factor influencing acceptance rates: documentation tasks achieve 82.1% acceptance compared to 66.1% for new features - a 16 percentage point gap that exceeds typical inter-agent variance for most tasks. OpenAI Codex achieves consistently high acceptance rates across all nine task categories (59.6%-88.6%), with stratified Chi-square tests confirming statistically significant advantages over other agents in several task categories. However, no single agent performs best across all task types: Claude Code leads in documentation (92.3%) and features (72.6%), while Cursor excels in fix tasks (80.4%).

cs.SE↗

arXiv2025

Empirical and Sustainability Aspects of Software Engineering Research in the Era of Large Language Models: A Reflection

Software Engineering (SE) research involving the use of Large Language Models (LLMs) has introduced several new challenges related to rigour in benchmarking, contamination, replicability, and sustainability. In this paper, we invite the research community to reflect on how these challenges are addressed in SE. Our results provide a structured overview of current LLM-based SE research at ICSE, highlighting both encouraging practices and persistent shortcomings. We conclude with recommendations to strengthen benchmarking rigour, improve replicability, and address the financial and environmental costs of LLM-based SE.

cs.SE↗

arXiv2025

GA4GC: Greener Agent for Greener Code via Multi-Objective Configuration Optimization

Coding agents powered by LLMs face critical sustainability and scalability challenges in industrial deployment, with single runs consuming over 100k tokens and incurring environmental costs that may exceed optimization benefits. This paper introduces GA4GC, the first framework to systematically optimize coding agent runtime (greener agent) and code performance (greener code) trade-offs by discovering Pareto-optimal agent hyperparameters and prompt templates. Evaluation on the SWE-Perf benchmark demonstrates up to 135x hypervolume improvement, reducing agent runtime by 37.7% while improving correctness. Our findings establish temperature as the most critical hyperparameter, and provide actionable strategies to balance agent sustainability with code optimization effectiveness in industrial deployment.

cs.SE↗

arXiv2025

Implementation of Worsey-Farin splines with applications to solution transfer

This work primarily focuses on providing full implementation details for Worsey-Farin (WF) spline interpolation over tetrahedral elements. While this spline space is not new and the theory has been covered in other works, there is a lack of explicit and comprehensive implementation details, which we hope to provide. In this paper, we also demonstrate the effectiveness of the WF-spline space through a simple target application: solution transfer. Moreover, we derive an error estimate for the WF spline-based, solution transfer process. We conduct numerical experiments quantifying the conservative nature and order of accuracy of the transfer process, and we present a qualitative evaluation of the visualization properties of the smoothed solution. Additionally, in our study of conservation, we demonstrate how adaptive numerical quadrature rules on the tetrahedron used in conjunction with global L2-projection can improve the conservation of the solution transfer process.

math.NA↗

arXiv2025

Solar Orbiter's 2024 Major Flare Campaigns: An Overview

Solar Orbiter conducted a series of flare-optimised observing campaigns in 2024 utilising the Major Flare Solar Orbiter Observing Plan (SOOP). Dedicated observations were performed during two distinct perihelia intervals in March/April and October, during which over 22 flares were observed, ranging from B- to M-class. These campaigns leveraged high-resolution and high-cadence observations from the mission's remote-sensing suite, including the High-Resolution EUV Imager (EUI/HRI_EUV), the Spectrometer/Telescope for Imaging X-rays (STIX), the Spectral Imaging of the Coronal Environment (SPICE) spectrometer, and the High Resolution Telescope of the Polarimetric and Helioseismic Imager (PHI/HRT), as well as coordinated ground-based and Earth-orbiting observations. EUI/HRI_EUV operating in short-exposure modes, provided two-second-cadence, non-saturated EUV images, revealing structures and dynamics on scales not previously observed. Simultaneously, STIX captured hard X-ray imaging and spectroscopy of accelerated electrons, while SPICE acquired EUV slit spectroscopy to probe chromospheric and coronal responses. Together, these observations offer an unprecedented view of magnetic reconnection, energy release, particle acceleration, and plasma heating across a broad range of temperatures and spatial scales. These campaigns have generated a rich dataset that will be the subject of numerous future studies addressing Solar Orbiter's top-level science goal: "How do solar eruptions produce energetic particle radiation that fills the heliosphere?". This paper presents the scientific motivations, operational planning, and observational strategies behind the 2024 flare campaigns, along with initial insights into the observed flares. We also discuss lessons learned for optimizing future Solar Orbiter Major Flare campaigns and provide a resource for researchers aiming to utilize these unique observations.

astro-ph.SR↗

arXiv2025

Exploring the Impact of Generative Artificial Intelligence in Education: A Thematic Analysis

The recent advancements in Generative Artificial intelligence (GenAI) technology have been transformative for the field of education. Large Language Models (LLMs) such as ChatGPT and Bard can be leveraged to automate boilerplate tasks, create content for personalised teaching, and handle repetitive tasks to allow more time for creative thinking. However, it is important to develop guidelines, policies, and assessment methods in the education sector to ensure the responsible integration of these tools. In this article, thematic analysis has been performed on seven essays obtained from professionals in the education sector to understand the advantages and pitfalls of using GenAI models such as ChatGPT and Bard in education. Exploratory Data Analysis (EDA) has been performed on the essays to extract further insights from the text. The study found several themes which highlight benefits and drawbacks of GenAI tools, as well as suggestions to overcome these limitations and ensure that students are using these tools in a responsible and ethical manner.

cs.AI↗

arXiv2024

Adaptive Self-Supervised Learning Strategies for Dynamic On-Device LLM Personalization

Large language models (LLMs) have revolutionized how we interact with technology, but their personalization to individual user preferences remains a significant challenge, particularly in on-device applications. Traditional methods often depend heavily on labeled datasets and can be resource-intensive. To address these issues, we present Adaptive Self-Supervised Learning Strategies (ASLS), which utilizes self-supervised learning techniques to personalize LLMs dynamically. The framework comprises a user profiling layer for collecting interaction data and a neural adaptation layer for real-time model fine-tuning. This innovative approach enables continuous learning from user feedback, allowing the model to generate responses that align closely with user-specific contexts. The adaptive mechanisms of ASLS minimize computational demands and enhance personalization efficiency. Experimental results across various user scenarios illustrate the superior performance of ASLS in boosting user engagement and satisfaction, highlighting its potential to redefine LLMs as highly responsive and context-aware systems on-device.

cs.CL↗

arXiv2024

Scalable Differential Privacy Mechanisms for Real-Time Machine Learning Applications

Large language models (LLMs) are increasingly integrated into real-time machine learning applications, where safeguarding user privacy is paramount. Traditional differential privacy mechanisms often struggle to balance privacy and accuracy, particularly in fast-changing environments with continuously flowing data. To address these issues, we introduce Scalable Differential Privacy (SDP), a framework tailored for real-time machine learning that emphasizes both robust privacy guarantees and enhanced model performance. SDP employs a hierarchical architecture to facilitate efficient noise aggregation across various learning agents. By integrating adaptive noise scheduling and gradient compression methods, our approach minimizes performance degradation while ensuring significant privacy protection. Extensive experiments on diverse datasets reveal that SDP maintains high accuracy levels while applying differential privacy effectively, showcasing its suitability for deployment in sensitive domains. This advancement points towards the potential for widespread adoption of privacy-preserving techniques in machine learning workflows.

cs.CR↗

arXiv2024

Minimal Attached Primes of Local Cohomology Modules of Binomial Edge Ideals of Block Graphs

We calculate the minimal attached primes of the local cohomology modules of the binomial edge ideals of block graphs. In particular, we obtain a combinatorial characterisation of which of these modules are non-vanishing. We also show that the main result of this paper follows from a recent result of Lax, Rinaldo, and Romeo (arXiv:2405.08671, Theorem 3.2), which was published independently during the writing of this paper. This provides a short alternative proof of our result.

math.AC↗

arXiv2023

Binomial Edge Ideals of Complements of Graphs of Girth at Least 5

We calculate the local cohomology modules of the binomial edge ideals of the complements of graphs of girth at least 5 using the tools introduced by \`Alvarez Montaner in arXiv:1901.08645. We then use this calculation to compute the depth, dimension, and regularity of these binomial edge ideals.

math.AC↗

arXiv2023

User-Centric Deployment of Automated Program Repair at Bloomberg

Automated program repair (APR) tools have unlocked the potential for the rapid rectification of codebase issues. However, to encourage wider adoption of program repair in practice, it is necessary to address the usability concerns related to generating irrelevant or out-of-context patches. When software engineers are presented with patches they deem uninteresting or unhelpful, they are burdened with more "noise" in their workflows and become less likely to engage with APR tools in future. This paper presents a novel approach to optimally time, target, and present auto-generated patches to software engineers. To achieve this, we designed, developed, and deployed a new tool dubbed B-Assist, which leverages GitHub's Suggested Changes interface to seamlessly integrate automated suggestions into active pull requests (PRs), as opposed to creating new, potentially distracting PRs. This strategy ensures that suggestions are not only timely, but also contextually relevant and delivered to engineers most familiar with the affected code. Evaluation among Bloomberg software engineers demonstrated their preference for this approach. From our user study, B-Assist's efficacy is evident, with the acceptance rate of patch suggestions being as high as 74.56%; engineers also found the suggestions valuable, giving usefulness ratings of at least 4 out of 5 in 78.2% of cases. Further, this paper sheds light on persisting usability challenges in APR and lays the groundwork for enhancing the user experience in future APR tools.

cs.SE↗