Papers

Concurrent multi-agent coding promises division of labor across modules, robustness through redundancy, and parallel exploration at the natural granularity of multi-file projects. Realtime collaborative editing protocols solve this coordination problem for human teams via Conflict-free Replicated Data Types (CRDTs), but the LLMs underneath generate one token at a time and existing multi-agent coding systems inherit this serial limit: they either sequence agents through phase handoffs or pool independent samples without coordination, and a single agent abandons up to half of hard tasks with a one-file stub-and-exit. AgentRoom is a realtime collaborative editing protocol for concurrent coding agents. Its runtime layer exposes file-level claim, status, and broadcast as MCP tools on a CRDT-merged shared filesystem. Five frontier coding-CLI models ran four backend coding tasks, with cross-language checks in Python DevBench and Rust+axum. For CLI-stable models, AgentRoom with 2 agents abandons fewer tasks than Solo and has less run-to-run variation. At matched-compute, one positive mean LLM-judge contrast puts AgentRoom over parallel-merge. The other contrast, a bundle probe, puts full AgentRoom above each partial case: an ordering rather than a percentage split. Coordination, not parallelism or CRDT-merge, bears the load.

Aligning deployed language models requires knowing when their outputs can be trusted, yet on-device models now ship to hundreds of millions of devices with no server-side moderation, and the configuration developers can actually deploy is rarely audited independently. We present a reproducible reliability audit of the developer-accessible on-device foundation model, framed as an oversight question: can a user or a resource-constrained developer tell when the model is wrong? Red-teaming it on calibration, confident confabulation on false-premise questions, and over-refusal of benign prompts, we find a task-asymmetric miscalibration: its guardrails fail in opposite directions across tasks (confabulating on 69% of false premises while refusing 18% of entirely benign inputs), atop a self-reported confidence that is saturated and non-discriminative (AUROC 0.47; ECE 70, worst among comparable small models). Crucially, confident-correct and confident-wrong outputs are surface-indistinguishable: a classifier over 15 user-visible features separates them at AUROC only 0.55 (equivalence-confirmed), leaving no signal for oversight at inference time. No cheap single-generation signal flags these failures ($\le$0.68 AUROC), whereas a black-box consistency wrapper requiring no model access recovers reliability (confident confabulation 75%$\to$3%; selective accuracy 43%$\to$83%) at a tunable cost. We contribute a model-agnostic audit protocol, a surface-indistinguishability test, and released code and frozen evaluation items as reusable infrastructure for auditing deployed models.

AI agents are increasingly used for simulation-driven engineering. Physical system modeling presents different requirements from general-purpose code generation in software engineering, because correctness depends not only on syntax and executability but also on physical consistency and scenario-dependent behavior. We study this challenge in Modelica, an equation-based modeling language in which a model may compile and simulate while still violating its intended physics or engineering requirements. Across successive revisions, an agent may lose track of requirements or rely on simulation evidence produced by an outdated candidate. To address this challenge, we present Pufibara, an agent harness that maintains persistent engineering state across revisions, associates execution and simulation evidence with the candidate that produced it, and makes submission an explicit agent action. To evaluate end-to-end Modelica agent workflows, we also propose a source-grounded method for constructing realistic and independently evaluable tasks. We use this method to build the 232-task Modelica Agent Workflow Benchmark, spanning Model Repair, Model Generation, and Model Tuning. Each submitted candidate is scored by a benchmark-owned evaluator outside the agent loop. We compare Pufibara with Claude Code as complete harnesses under two matched large language model (LLM) backends. With DeepSeek v4 Flash, Pufibara passes 202 tasks, compared with 185 for Claude Code. With Claude Sonnet 5, Pufibara passes 202 tasks, compared with 187 for Claude Code. Under the repository-reported token accounting, Pufibara records 76.4%-82.5% lower logical-token totals. Its sequential runtime is 6.1%-58.4% lower. These findings show that, even under matched LLM backends, complete agent harnesses can differ substantially in both task success and resource use for physical system modeling.

Agent harnesses record a failed tool call and its error message in the transcript and ask the model to continue, on the assumption that the error is corrective information. We measure whether it is. Defining the corrective gain of a failure record as the change in log-probability of re-emitting the action that just failed, we find the gain is negative for every instruction-tuned model we tested (6 checkpoints, 135M-1.7B, 4 families) in two environments: simulated tool calling and MBPP program repair. Normalised by action length the effect is about -1.03 nats per action token, a factor of 2.8 in the odds of each token, and holds on 90%-100% of individual items, not only on average. Over a fixed candidate set the probability of repeating the failed call rises from 0.06 to 0.54, and greedy decoding reproduces it token for token on 19% of items after the failure versus 0% before. Counterfactuals pairing the same call with a failure message, a success message, or a neutral acknowledgement separate two effects: the failed call's surface form accounts for 83% of the damage, while the semantic contribution of marking it failed is small and inconsistent in sign across environments. The problem is in the harness, not the model's grasp of error messages, and that predicts which remedies work. Replacing the verbatim call with a runtime-generated description of the failure removes 76% of the inversion at no token cost, and making previously-failed strings unreachable at the decoder acts on the same term. Two plausible remedies do not: an explicit "do not repeat" instruction leaves the measured quantity where it was, and deleting the failed attempt to retry from a clean context, the standard prescription for context contamination, is the worst harness we measured for repetition, because it restores the context that produced the failure. The study runs end to end on a CPU; all artefacts are released.

Microservice-based systems are modular and adaptable, but their distributed structure makes their timing behavior difficult to analyze and guarantee. Because latency emerges from interactions among service dependencies, shared resources, and coordinating middleware, local timing disturbances can propagate into system-level effects. This survey examines the sources, propagation mechanisms, and observable impacts of timing variability in microservice-based systems, using software-defined vehicles as a motivating example. It compares major classes of timing models by their assumptions about workload stability, execution structure, resource sharing, observability, and guarantee admissibility, and summarizes their limitations within microservice-based software-defined vehicles.

Modern software systems accumulate technical debt over decades of development, which makes migration expensive and largely manual. As coding agents become increasingly capable at bug fixing, can they autonomously perform such migrations? Existing benchmarks cannot answer this question because they evaluate only behavioural correctness, not whether the migration actually occurred. This leads an easy hack: agents copy the original implementation to make tests pass. We call this Blindness. To address this problem, we introduce SWE Refactor Bench, a benchmark comprising 20 whole-repository migrations, covering 4 kinds of technical debt. A three-stage evaluation protocol measures both migration completeness and behavioural correctness. (1) Migration Audit verifies that the migration occurred. (2) Behavioural Tests measure correctness with a fixed test suite. (3) Agentic Verification uses 6 independent coding agents to generate targeted tests for hidden behavioural differences. Across 520 runs from 8 frontier models and 26 model-effort configurations, only 28 of 520 runs ($5.4\%$) pass all three stages, 13 of the 20 tasks receive no accepted solution, and the best model (claude-opus-5) scores $47.0/100$. Migration completeness and behavioural correctness are distinct abilities: a few runs preserve behaviour by skipping the migration and are stopped at Migration Audit; most attempt it and break behaviour, and are stopped at Behavioural Tests. Agents cannot deliver a perfect migration: among the 340 runs that pass Migration Audit, $58\%$ reach $99\%$ of the fixed checks, yet only $26\%$ reach $100\%$. Agent capability differs across migration categories: agents score $31.4$ on build toolchain rewrites but only $5.6$ on language rewrites. Together, these findings position SWE Refactor Bench as a rigorous testbed for developing coding agents for reliable whole-repository migrations.

Language models are sequential processors, but long-horizon agency requires external information and computation beyond model weights and active context. Prime Agent is an open-source harness for long-horizon evaluation and coding-agent workflows. A persistent IPython REPL follows the Recursive Language Model abstraction for programmatic context processing and test-time compute, while Continual Harness preserves histories, memories, skills, prompts, and subagent specifications across trajectories. Recursive subagents coordinate through direct agent-to-agent communication, and the Agents View lets humans inspect and manage daemon-backed sessions. Prime Agent standardizes execution, recovery, verification, and resource accounting while leaving strategy construction to the model. This low-friction, expressive membrane prevents harness failures from becoming model failures and pushes measurement toward the model's true maximal underlying capability. Prime Agent raises ARC-AGI-3 RHAE Best@1 from 30% to 95.5% and matches or exceeds native and popular harnesses across long-context coding, GPU-kernel generation, emulator construction, and autonomous nanoGPT speedruns. On Factorio, we find refinement allows for continuous technology progression and dedicated subagents enable parallelized work. Code is available at https://github.com/PrimeIntellect-ai/prime-agent.

Finding people from a natural-language description (“ML engineers transitioning to research roles in biotech”) is increasingly delegated to LLM agents and framed as information retrieval. We argue that it is fundamentally a requirements engineering task: such a request is an under-determined requirement with implicit constraints, many valid answers, and no acceptance criterion, so useful answers require eliciting, validating, and verifying the requirement before search can matter. We present \sys{}, to our knowledge the first interactive, requirements-driven candidate-sourcing agent (it elicits, validates, retrieves, and verifies a vague people-request into a justified slate through bounded elicitation, workflow templates, a two-stage commit protocol, and bidirectional termination guards) and \bench{}, a benchmark that runs the requirements lifecycle (criteria-anchored validation, multi-model evidence-grounded oracle construction, and cost-aware verification). Across $21$ systems and all $691$ requirements, \sys{} dominates breadth ($100%$ coverage at $2.5\times$ the yield) and is near-orthogonal to the field, with $90%$ of the people it returns are surfaced by none of $20$ strong LLM-plus-web baselines combined. Beyond breadth, an evidence-grounded judging of every system shows \sys{} recalls the most relevant real people: $0.241$ of the union pool, $1.9\times$ the next system, with a bootstrap $95%$ interval disjoint from every baseline. \sys{} is thus the strongest sourcing engine (the deepest real, reachable candidate pool), while precision-ranking LLMs serve as~complementary verifiers.

Recent LLM-agent systems make conflicting design bets: decompose work across many narrow agents, or use one strong tool-using agent. This pilot studies that choice on bounded cross-border VAT determination with reverse charge, where every case has an oracle label and each intermediate decision is independently scoreable. We hold the activity surface fixed (subtasks, tools, I/O schemas, validation checks, orchestrator, base model, and merge policy) and vary only the assignment of subtasks to workers across four orchestrated configurations, from one wide worker to five narrow ones, against S0, a tuned no-orchestrator single agent, with a deterministic rule engine as oracle. The program spans 4,400 runs: a 40-case, five-repeat main sweep, matched-token arms separating prompt-budget from agent-count effects, and three failure-injection arms, all judged against pre-registered falsification criteria. The two intermediate configurations lead on accuracy (0.830, against endpoints at 0.720 and 0.770) but miss the pre-stated bar against the fine endpoint, so the intermediate-optimum hypothesis remains unsupported at pilot scale. The single agent does not Pareto-dominate the orchestrated set. The matched-token criterion fires: the budget-matched single agent lands 6.5 points below the leader, but the interval includes zero, so any advantage is consistent with a prompt-budget explanation. Under injection, availability faults are absorbed at every granularity, with wide-scope restart over-recovering its baseline by +0.160, while one schema-conforming hallucinated record degrades every configuration and inverts the ordering, hitting fragmented configurations hardest. The contribution is a bounded, preregistered pilot heuristic for right-sizing decomposition (place one partition boundary at the dependency-layer midpoint), released with oracle, dataset, harness, raw traces, and analysis pipeline.

Developers use automated Move refactorings to improve the modular structure of source code and the assignment of responsibilities. Class- and method-level Move refactorings are automated in modern IDEs, but statement- and expression-level moves that adjust method boundaries remain largely unautomated. We formalize five variants of Move Statement refactoring as preconditions and steps grounded in four basic conditions covering data reachability, execution count, side effects, and syntactic constraints required for compilation, of which all but the side-effect condition are checked statically. Combined with existing techniques, this also yields finer-grained moves of expressions and partial expressions. We further refine the formalization iteratively against a real project, deriving twenty additional preconditions and steps that handle Java syntactic diversity in practice. We evaluate applicability and compilability on ten projects, and behavior preservation in a case study on one of them: Move Statement refactorings yield compilable code in 93.3-97.0% of applicable cases, and the case study shows that the observed behavioral changes stem from side-effect reordering left to developer judgment, not from defects in the statically checked conditions.

Reproduction test generation, producing a failing-then-passing test that captures a reported bug, is a critical step in automated software engineering. Existing agentic methods treat this as a monolithic loop, despite the task inherently comprising two subtasks of distinct nature: diagnosing the root cause and writing a fail-to-pass test. Without explicit separation, the agent faces a compound objective with underspecified intermediate goals, leading to goal drift. We propose DPIAgent, a structured agentic framework built on three principles, Divide, Protocol, Isolate (DPI), that mitigates compound-objective ambiguity and goal drift: it Divides the task into single-objective phases of defect exploration and test generation; enforces a handoff Protocol that records the diagnosis and test plan, preventing context loss; and Isolates each phase's action space by tailoring the toolset to its task, preventing irrelevant tools from misleading execution. On SWT-Bench Verified, DPIAgent outperforms seven baselines across three backbone LLMs. With DPI alone it reaches 81.76% success rate on GPT-5, the highest reported among open-source methods, gaining up to 11.88 points over the strongest baseline on GPT-5-Mini; adding test selection further raises it to 86.17%. Our analysis shows that architectural structure and backbone capability are complementary axes rather than substitutes, demonstrating DPI's generalizability across model classes.

Large Language Models (LLMs) agents deployed in vehicles must satisfy a written operating policy on every turn: a single hallucinated identifier, omitted mandatory side-effect, or premature completion claim fails the task. We present AgentGuardUtil, our entry to CAR-bench Track~1, which treats the AI planer (LLM) as a fallible proposer inside a grounded verify-and-revise loop. Its core novelty is a runtime policy compiler: the natural-language policy shipped with each conversation is compiled, once per policy, into typed machine-checkable rules, a subset of which receive an executable form. A deterministic obligation engine interprets these rules against live tool results and the simulated post-write state of the draft itself, emitting the exact remedial calls with computed arguments rather than natural-language reminders. Around this engine, 25 deterministic gates (identifier provenance, schema and enum validity, gather-before-act, confirmation and future-time protocols) and an LLM critic produce tiered findings that drive a bounded revision loop tuned for the pass k metric.

We investigate the software security and maintenance practices adopted by stakeholders in the TianoCore community and identify opportunities to improve firmware development workflows. We conduct a survey and a limited interview study with participants representing independent firmware vendors, original equipment manufacturers, security experts, firmware developers, and academic researchers. This open-source development community maintains a reference implementation for the core of the UEFI firmware. We highlight important gaps in the current state of firmware development within the TianoCore ecosystem and identify key areas in which improved security practices, greater adoption of memory-safe technologies, and increased automation of manual processes could strengthen the maintenance and security of the UEFI firmware.

We propose a novel approach to bug triage in the TianoCore open-source UEFI firmware development ecosystem. This integrated approach, called TianoForge, deploys the state of the art in artificial intelligence, specifically machine learning, to enable automated bug triage. This includes invalid bug report detection, duplicate bug report detection, bug report prioritization, and bug report assignment. We use various Generative Pretrained Transformer (GPT) Large Language Models (LLMs) with and without Retrieval Augmented Generation (RAG) to automate these tasks. Given the crucial role of bug triage in software maintenance and the huge number of untriaged issues in the TianoCore community, in particular, their primary project, EDK II, we expect a significant impact on the efficiency of TianoCore software maintenance processes, primarily bug triage and resolution. Our experimental study shows that TianoForge reduces the average bug triage time from around 11 days to approximately 7 minutes, which is a 99.95% reduction.

FAIR (Findable, Accessible, Interoperable, and Reusable) metadata is essential for the discovery, interoperability, and reuse of scientific research assets. However, creating and maintaining FAIR metadata remains largely manual, making the process time-consuming for heterogeneous research artifacts generated throughout the research lifecycle. Existing approaches primarily generate metadata at publication time, missing opportunities to capture contextual information as it becomes available. To address this limitation, we present LLMCrater, a lifecycle-aware metadata generation framework that combines Large Language Models (LLMs) with stage-specific RO-Crate metadata profiles. The framework progressively enriches metadata across four research lifecycle stages (Design, Development, Deployment, and Execution & Provenance) while remaining compatible with RO-Crate~1.1 and EOSC metadata recommendations. It automatically extracts metadata from heterogeneous artifacts, generates and validates machine-actionable RO-Crates, and supports publication to FAIR repositories and PID services (e.g., Zenodo). We demonstrate the approach using two representative use cases: a 5G experimentation environment within SLICES-RI and an experiment on GreenDIGIT's EcoJupyter platform. Results show that LLMCrater progressively enriches metadata throughout the research lifecycle and generates valid RO-Crates conforming to the RO-Crate~1.1 specification.

Legacy scientific applications remain valuable research assets but are often tightly coupled to project-specific execution environments, limiting their reuse, reproducibility, and deployment within modern scientific workflow systems and cloud-native Virtual Research Environments (VREs). Existing migration approaches primarily target individual artifacts, such as notebooks or containers, and provide limited support for systematically transforming heterogeneous legacy applications into reusable cloud-native workflows. This paper presents an AI-assisted migration framework that combines the Reference Model of Open Distributed Processing (RM-ODP)-guided architectural analysis, Large Language Models (LLMs), and Design Structure Matrix (DSM) analysis. The framework first uses RM-ODP to guide an LLM in identifying reusable workflow components, their interfaces, and execution dependencies from heterogeneous legacy applications. The resulting workflow structure is then iteratively evaluated and refined using DSM analysis. Finally, an LLM-based workflow generator implements the validated workflow components and produces containerized execution environments and executable workflow definitions for deployment in cloud-native workflow systems, including VREs. The framework was evaluated on two legacy scientific applications from different scientific domains. In both cases, the applications were successfully transformed into reusable cloud-native workflows while preserving their original functionality, demonstrating the feasibility of the proposed approach for modernizing legacy scientific software.

Research software is increasingly central to scientific workflows, yet it is often developed by researchers with limited software engineering expertise. This can lead to quality issues that hinder maintainability, reproducibility, reuse, and sustainability. Existing static analysis tools can identify such issues, but their outputs often require expert interpretation and provide limited support for translating quality assessments into actionable improvements. To address this gap, we propose a lifecycle-aware framework that integrates quantitative software quality assessment with Large Language Model (LLM)-based code refinement. The framework comprises two stages. First, a lifecycle-aware Quality Model is developed from established software quality standards and practitioner requirements. The model defines five quality dimensions and 25 candidate metrics, of which 14 are operationalized using existing analysis tools and custom measurements. Second, the resulting quality diagnostics are used as structured feedback within an iterative LLM-based refinement process, enabling generated improvements to be repeatedly reassessed against the Quality Model. We evaluate the framework on notebook-centric research software using multiple LLMs and compare iterative structured feedback with single-step feedback and unstructured prompting. The results show improvements in specific quality attributes, particularly code duplication and structural quality, while also revealing trade-offs among maintainability, code size, documentation, and complexity. These findings demonstrate the potential of metric-driven LLM feedback for research software quality improvement while highlighting its inherently multi-objective nature \footnote{The source code and experimental data are publicly available at https://github.com/QCDIS/Software_Quality_Control_LLM . }

Digital Twin (DT) ecosystems integrate heterogeneous computational models to represent complex systems under evolving, purpose-specific objectives. Systematic reuse of existing high-quality models and datasets is essential for scalable DT development, yet is constrained by heterogeneity in semantic intent, data structures, behavioral interfaces, and execution environments. As a result, integration becomes a cross-model, cross-view consistency problem that is hard to predict, quantify, and compare across design choices. Existing standards and integration platforms address these concerns separately, offering limited support for structured, purpose-aware compatibility assessment and early feasibility analysis when models are reused under new DT objectives. This paper introduces a multi-viewpoint integration modeling framework grounded in the Reference Model of Open Distributed Processing (RM-ODP). The framework structures integration-relevant knowledge across domain, information, computational, engineering, and technology viewpoints, representing cross-view dependencies as explicit, machine-actionable metadata. It comprises (i) a viewpoint-structured Model Metamodel for systematic model description and discovery, and (ii) a pattern-aware Mismatch Detector that operationalizes cross-view compatibility constraints via integration patterns, combining deterministic rule generation with Large Language Model (LLM)-assisted reasoning. This enables systematic identification of semantic, informational, and runtime inconsistencies and supports reasoning about integration feasibility and effort before implementation. Expert validation and an environmental modeling case study show that the approach enables structured compatibility reasoning, improves transparency of integration assumptions, strengthens cross-view interoperability, and supports scalable reuse in heterogeneous DT ecosystems.

Kubernetes incident triage requires correlating signals from metrics, logs, container state, and messaging systems across multiple monitoring tools, a fragmented workflow that slows diagnosis and contributes to alert fatigue. Large language models (LLMs) have shown promise for automated root cause analysis (RCA), but existing systems rely on custom, system-specific data access layers that cannot be reused across organisations. We present ARGUS, an MCP-grounded RCA assistant that connects a commercial LLM to live Kubernetes observability data through standardised MCP servers covering Kubernetes state, Prometheus metrics, Loki logs, and NATS messaging, and delivers structured diagnostic summaries inside the Slack incident channel where on-call engineers already work. We conduct a preliminary evaluation of ARGUS using three complementary methods: controlled fault injection across ten Kubernetes incident scenarios, rubric-based scoring of the resulting RCA summaries on three dimensions, and semi-structured interviews with six on-call engineers at an industrial partner. ARGUS named the correct root cause in all ten scenarios with an aggregate MCP success ratio of 0.91. Practitioners trusted the diagnostic output but consistently expressed scepticism toward the recommended fixes. Our central finding is a diagnostic/prescriptive asymmetry: ARGUS reliably identifies what went wrong, but is perceived as less reliable or trustworthy at specifying what to do next. This pattern can be observed across all three evaluation methods, and has important implications for future autonomous agentic incident handling systems.

LLM agents remain unreliable on long-horizon tasks, where small local failures can compound over extended interactions and lead to overall task failure. Although external harnesses can substantially improve robustness, harness design remains a manual and expensive process that requires searching over a large space of prompts, tool configurations, and control logic. We propose AutoSaddler, an automatic harness optimization framework that formulates harness improvement as an offline learning problem and iteratively updates the harness using failure signals from mini-batches. AutoSaddler combines failure-trace diagnosis, structured patch generation that treats the harness as code, and validation-based update selection. Experiments on GAIA2, SWE-Bench Pro, and Terminal-Bench 2.0 show that AutoSaddler substantially improves agent performance over the corresponding base harnesses, achieving gains of 9.0, 9.6, and 10.0 percentage points, respectively. Ablation studies further suggest that effective harness optimization benefits from three ingredients: deep debugging rather than shallow reflection, targeted modifications rather than unconstrained editing, and generalization-aware selection rather than trajectory-specific repair. Together, these results suggest that automatic harness optimization is a promising path toward more performant and reliable agent systems.

Automated manuscript pipelines often regenerate an entire section to repair a local defect, allowing unrelated metrics and citations to change even when the resulting PDF still builds. PatchWrite instead constrains how candidate edits become committed manuscript states: it reuses bounded EDIT N M editing and rollback, but tightens compilation acceptance with fatal-log checks and adds evidence locks that require every cited key and experimental numeric token to be attested by a reference registry or experimental log. Candidates that fail either check are rejected and the previous HEAD is retained. On a 24-manuscript x 8-fault oracle stress test (768 jobs, evenly split between compile-breaking and content-only faults), whole-slot rewriting mutated an unrelated "12-layer" line in every case (0/192 preserved; numeric Jaccard 0.6667), whereas PatchWrite preserved it in 192/192 cases. Removing the compile gate reduced acceptance to 0, while removing the evidence gate allowed a hallucinated citation to pass. The same pattern held across all eight faults. To test the protocol with generation rather than oracle edits, we reran the 192 jobs with the writer model proposing the edits. The model's candidates were accepted in 75% of cases; nearly all rejections came from one reproducible failure mode in which the model attempted to delete a line using an empty replacement unsupported by the current grammar. Every accepted candidate passed both gates, and 93.75% fixed the injected fault; the remaining cases involved a technically valid but sentence-inappropriate citation and one markup-changing near-miss. In a blind evaluation of sixteen PDF pairs, both raters preferred PatchWrite for preserving lab-grounded facts (C1 Likert 5.0 vs. 2.0), while rating prose quality nearly identically. Logs from 193 in-product drafting tasks show the same classes of failures occurring in practice.

Fairness is increasingly recognised as a first-class requirement in socio-technical processes, where interactions among human actors, software systems, and AI technologies may lead to unfair outcomes in decision-making workflows. If left unaddressed, fairness hazards may accumulate and reinforce systemic bias, highlighting the need to engineer fairness proactively. Despite growing interest in fairness-aware systems, systematic methods for identifying fairness hazards in socio-technical processes and deriving requirements-level mitigations remain limited. To support fairness-by-design during requirements engineering (RE), Fairness Hazard Analysis (FHA) is introduced as a methodology for systematically identifying, analysing, and mitigating fairness hazards. FHA is first assessed through a proof-of-concept validation conducted via two focus groups. Then, a qualitative multiple-case study involving two organisations examines its applicability in real-world settings. The proof-of-concept validation highlighted the benefits derived from the structured nature of the method, and suggested the need to include iterative, dialogic reflection with domain experts. In the multiple case-study where FHA was applied, the practitioners involved were positively impressed by the results and confirmed the relevance of the identified fairness hazards (spanning up to 27% of the process elements), as well as the appropriateness of most of the proposed mitigations, while noting that contextual factors might hinder their implementation. The evaluation also highlighted mitigation patterns, such as independent review and collective decision-making, which can be transferred to different organisations. This paper contributes a structured and empirically validated methodology for integrating fairness considerations in RE and preventing systemic bias in socio-technical processes.

Large Language Models (LLMs) have demonstrated remarkable capabilities in code generation, yet their performance degrades significantly on low-resource Hardware Description Languages such as Verilog. While multi-candidate sampling improves the likelihood of generating correct solutions, au-tomatically selecting the optimal candidate remains an open challenge. Through a systematic empirical study across nine models and two benchmarks, we identify two critical limitations:(1) existing execution-based reranking methods, which rely on testbench pass/fail outcomes, exhibit poor domain transferability due to low-quality generated testbenches; and (2) LLM-as-a-Judge suffers from reasoning hallucination, producing incon-sistent judgments for execution-equivalent code. These findings reveal two signal types with orthogonal errors: execution signals(deterministic but testbench coverage limited)and reasoning signals (semantically rich but hallucination-prone). Their orthog-onality suggests combining the two signals, yet in our experiments letting the reasoner directly observe execution results merely anchors its judgments on test outcomes; we therefore acquire the two signals independently and fuse them only at the decision stage. Based on these insights, we propose EAHC, an Execution-Anchored Hallucination Calibration reranking framework that anchors reasoning judgments to execution behavior so that execution-equivalent candidates receive consistent scores, which implements a dual-channel architecture: EAHC-R, a 4B reasoning discriminator; and EAHC-T, a testbench generator leveraging RAG for execution verification.

Agentic AI is a fascinating new tool for software development. It is a huge step forward compared to "conventional" AI assisted coding, which in turn was a considerable breakthrough earlier. AI support through LLMs is a young and very fast-moving field. The "conventional" (non-agentic) flavor became useful and productive in early 2025 (around 18 months ago) and the agentic flavor followed in fall 2025 (approximately 9 months ago). Besides all its benefits and potential, it also carries some fundamental risks for IT security. And the agentic approach added very severe risks while making others much more dangerous. With all the motivation to explore this fascinating new tool we should not ignore the risks but actively address them. We present (I) an assessment of the IT security risks, (II) a concept for mitigating them without breaking its benefits, and (III) an overview about an implementation of our concept. In this very dynamic field this is likely not the final and once-and-for-all answer to the identified issues but still a substantial step forward in responsible usage of Agentic AI for software development. It should also be a contribution to the community to allow early and eager evaluation of the potential of agentic AI for software development without actually suffering from its implied IT security risks.

Large language models (LLMs) rely on tool calling as a fundamental agent capability, enabling them to invoke external systems and complete tasks beyond text generation. However, clean end-to-end (E2E) success cannot identify where a tool-use failure originates or how it propagates through a call. We introduce ToolRobustBench, a stage-wise diagnostic benchmark for tool-calling agents, where a tool-calling agent is an LLM system that selects a tool, supplies structured arguments, and interprets its returned feedback. ToolRobustBench aligns four perturbation families with the tool-use pipeline: tool-interface, user-intent, tool-output/observation, and runtime-environment perturbations. It attributes failures to tool selection, schema grounding, argument binding, tool-output/runtime-feedback handling, and E2E task success. Experiments on 15,456 single-family instances across 7 models, 16 sampled local tools, 4 perturbation families, and 14 subtypes show high but non-uniform clean performance and substantial robustness degradation, with tool-output/observation perturbation the dominant bottleneck. Mixed-family experiments reveal non-additive failure patterns that are not explained by isolated single-family results. Thus, ToolRobustBench provides a deterministic and cascade-aware benchmark for diagnosing robustness beyond clean tool-calling accuracy;

Process supervision has improved mathematical reasoning, where intermediate steps are naturally expressed as chains of thought. In code generation, however, process supervision remains underexplored because there is no standard notion of a step. Supervision can target lines, reasoning traces, or program states, making it unclear what to label and optimize. We propose STEP-KTODER, a framework for code preference optimization that defines steps as module-level functions in decomposed multi-function programs and assigns binary correctness labels via automatically generated unit tests. Our method provides a code-specific instantiation of stepwise KTO, combining function-level process supervision with outcome-level feedback on the full program. We evaluate on HumanEval(+), MBPP(+), BigCodeBench, and LiveCodeBench, showing that STEP-KTODER improves over outcome-only KTO and DPO. Further analysis shows that execution-based labels are essential: LLM-as-a-judge annotations systematically over-predict function failures, corrupt positive step labels, and degrade downstream preference optimization. Code is available at: https://github.com/inechnech/STEP-KTODER.

Large language models (LLMs) have shown substantial promise for high-level synthesis (HLS) code generation, but most existing approaches validate only simulation or synthesis results. Because of timing and place-and-route constraints, HLS code that passes simulation and synthesis may still fail to produce deployable, runnable designs on real FPGA platforms. Moreover, the lack of public benchmarks has limited many evaluations to small, self-curated test suites. We propose FPGAgent, a multi-agent framework tailored to real FPGA environments for autonomous HLS coding with end-to-end executability validation. To the best of our knowledge, FPGAgent is the first task-specification-to-executable HLS generation framework experimentally validated on a well-established benchmark. Given a natural-language task specification, FPGAgent injects HLS-specific knowledge and employs evolutionary search to iteratively derive reliable HLS kernel implementations. It then generates a C++ validation program to verify functional correctness, diagnoses potential defects, and guides targeted repairs. Finally, it synthesizes host code for compilation and board-level execution on FPGA hardware. We comprehensively evaluate FPGAgent with five established LLMs on HLS-Eval, a benchmark containing 78 tasks across multiple domains, and verify board-level executability on a real FPGA platform. Compared with existing baselines, FPGAgent improves the synthesizable rate by 16.9% on average, executability by 26.7%, and functional correctness by 30.6%. These results show that FPGAgent substantially improves the practical usability of LLM-based HLS generation and demonstrates the value of end-to-end validation.

A tool call can be perfectly valid yet still leave an autonomous agent unable to determine what to do next. For example, if an external effect commits but its response is lost, committed and uncommitted states may become indistinguishable to the agent even though they require different continuation actions. We study this gap between callability and operability: whether a tool interface exposes the action-relevant state and semantics needed for an agent to continue safely under operational uncertainty. We operationalize tool operability through Agent-First Tooling (AFT), a set of interface mechanisms spanning selective capability discovery, execution lifecycle and recovery, explicit external-effect semantics, machine-readable results, and postcondition verification. We introduce AFT-Bench, a controlled interface-intervention framework that holds the task, backend, initial state, injected failure, agent, and language model fixed while varying the interface exposed to the agent.

LLMs are increasingly used for code generation, yet they frequently hallucinate non-existent software packages, creating exploitable entry points into the software supply chain. We make four contributions to this problem. First, we show that prior evaluation methodologies systematically inflate hallucination rates by misclassifying standard-library modules as hallucinations in some languages. For Python, the overestimation reaches 9.4 percentage points. Second, we evaluate seven inference-time defenses for mitigating package hallucinations, including five guided decoding strategies (Greedy, Contrastive, DoLa, Nudging, and Active Layer-Contrastive Decoding), an iterative self-refinement approach (Self-Refine), and a Retrieval-Augmented Generation (RAG)-based defense.. Across eight models spanning five families and four programming languages (Python, JavaScript, Ruby, Rust), RAG reduces the package hallucination rate (PHR) in 18 of 32 model–language configurations. Third, we introduce Package Utility (PU) to assess whether defenses preserve valid and task-relevant recommendations. Among strategies evaluated, Greedy decoding provides the strongest average mitigation–utility trade-off. Fourth, we stress-test all strategies under adversarial prompts seeded with fabricated package names and find that PHR surges by up to 45 percentage points relative to standard prompts, with Ruby consistently the most vulnerable language (80.9–95.2%). Under adversarial conditions, RAG and Self-Refine outperform all decoding-only strategies, indicating that robust defense requires either external grounding or iterative self-verification when prompts are actively hostile. Our results recast package hallucination as both a measurement problem and a decoding-time control problem, and they demonstrate that the choice of defense must be matched to the threat model and recommendation utility.

Copying a function from a chat window into an editor takes less than a second. For many uses of AI coding tools, that speed is the point; in settings such as programming education, code review, and security-sensitive development, it can also be the problem. This paper frames copy-paste as an AI code handoff problem: the moment model-generated text crosses from a conversational context into executable or committed software is a design boundary that current tools leave largely unmanaged. We argue that AI coding assistants should not only be evaluated by the code they generate, but also by how they mediate the transfer of that code into software artifacts. We propose soft barriers as one class of handoff-aware mechanisms. Soft barriers preserve access to AI assistance while making unexamined transfer less frictionless. As an initial technical probe, we instantiate this idea using Unicode output perturbations that preserve visual readability but disrupt naive copy-paste execution. We introduce Copy-Paste Resistance (CPR), the fraction of functionally correct clean solutions that become syntactically invalid after perturbation. Across HumanEval and MBPP with four LLMs and four perturbation families, we find that output-level barriers can achieve high copy-paste resistance, but their effectiveness is highly model- and task-dependent. An exploratory pilot with 18 participants provides early evidence that soft barriers can shift users from direct transfer toward editing and reconstruction. We do not present Unicode perturbations as a deployment-ready solution; rather, we use them as a minimal probe for a broader research agenda on practical, transparent, and policy-aware AI code handoff.