UML state machine design is a critical process in software engineering. Traditionally, state machines are manually crafted by experienced engineers based on natural language requirements-a time-consuming and error-prone procedure. Many automated approaches exist but they require structured NL requirements. In this paper, we investigate the capabilities of current Large Language Models to fully automate UML state machine generation via specialized State Machine Frameworks (SMFs) from non-structured NL requirements. We evaluate two types of state-of-the-art LLMs using single-step and multi-step prompting approaches: a non-reasoning LLM GPT-4o and a reasoning-focused LLM Claude 3.5 Sonnet, and introduce a novel Hybrid Approach that uses the output from a Single-Prompt Baseline as an initial draft state machine, which is then refined through an SMF. In our study, two distinct SMFs are developed based on human approaches: (i) a Structure-Driven SMF, in which state machine components (states, transitions, guards, actions, etc.) are generated in sequential steps, and (ii) an Event-Driven SMF, where identified events iteratively guide state machine construction. Our experiments indicate that while LLMs demonstrate a promising ability to generate state machine models from the Single-Prompt Baseline (e.g., F1-scores of 0.90 for states and 0.75 for transitions using Claude 3.5 Sonnet), their performance is not yet fully sufficient for a fully automated solution (e.g., F1-scores of 0.23 for guards and 0.00 for actions for GPT-4o). Our proposed Hybrid Approach improves the performance of the non-reasoning LLM (GPT-4o) to a similar level as the reasoning LLM (Claude 3.5 Sonnet) but does not further improve the reasoning LLM. Our evaluation highlights both the potential and the limitations of current LLMs for automated state machine design, providing a baseline for future research in this domain.

Large Language Models (LLMs) have made significant contributions to software engineering, particularly in the field of code generation, demonstrating the ability to produce functionally correct code snippets. The development of these models involves multiple stages, including pre-training on large amounts of source code data and aligning them with human preferences using various techniques. However, their development process often neglects foundational software engineering (SE) practices and principles. Specifically, LLMs receive limited exposure to core SE concepts during training, such as modularity, single responsibility, cohesion, and coupling. As a result, the generated code may lack the properties critical for building maintainable, extensible, and robust software systems. This vision paper advocates integrating SE knowledge directly into LLMs to enhance their capability to generate code and SE artifacts that adhere to and align with established best practices. We propose a new direction for LLMs to move beyond their current focus on functional accuracy toward producing robust, maintainable software. To assess how well LLMs internalize SE knowledge, we propose adopting Bloom's Taxonomy as a comprehensive assessment framework, offering a structured alternative to limited evaluation methods such as probing. By embedding software engineering principles, next-generation LLMs can leverage decades of software engineering knowledge and transform software development with reliable, high-quality generative capabilities.

Large language models (LLMs) are increasingly used to generate software artifacts across many software engineering (SE) tasks, yet ensuring the semantic validity of these artifacts remains a fundamental challenge. Existing constrained decoding techniques can enforce syntactic correctness and, in some cases, specific semantic rules, but lack a general representation that bridges LLM-generated text with the reasoning required for semantic validation in SE. In this paper, we propose projectional decoding, a novel conceptual framework that integrates domain semantics directly into the generation process by maintaining, alongside text, a partial graph model as the primary artifact representation throughout generation. This abstract representation enables incremental semantic validation by explicitly capturing uncertainty and natively supporting error detection, while guiding generation toward semantically valid outputs with provable guarantees. We present preliminary results on a program generation task which demonstrate the potential of this approach to improve the semantic validity of LLM-generated artifacts. We also discuss how projectional decoding can enable verifiable automation with LLMs across various SE activities.

State-of-the-art Large Language Models (LLMs) excel in code generation at the function level. However, the output quality significantly declines when scaling to repository-level systems. Current workflows relying only on natural language prompts suffer from inherent ambiguity and a lack of verifiability. To address this, we propose structured spec-driven engineering (SSDE), a paradigm that leverages structured artifacts to guide LLM generation. We argue that structured specifications as LLM inputs make high-quality, repository-level code generation a tangible goal, while at the same time offering superior verifiability, leading to significant potential for improvement. We first investigate the feasibility of this vision through a pilot study generating Model-View-Controller (MVC) business logic for three software systems using five LLMs, and then highlight the potential, challenges, and future roadmap for SSDE.

Graph model generation from natural language requirements is an essential task in software engineering, for which large language models (LLMs) have become increasingly popular. A key challenge is ensuring that the generated graph models are consistent with domain-specific well-formed constraints. LLM-generated graphs are often partially correct due to inconsistency with the constraints, limiting their practical usage. To address this, we propose a novel abstraction-concretization framework motivated by self-consistency for generating consistent models. Our approach first abstracts candidate models into a probabilistic partial model and then concretizes this abstraction into a consistent graph model. Preliminary evaluations on taxonomy generation demonstrate that our method significantly enhances both the consistency and quality of generated graph models.

Graph model generation from natural language description is an important task with many applications in software engineering. With the rise of large language models (LLMs), there is a growing interest in using LLMs for graph model generation. Nevertheless, LLM-based graph model generation typically produces partially correct models that suffer from three main issues: (1) syntax violations: the generated model may not adhere to the syntax defined by its metamodel, (2) constraint inconsistencies: the structure of the model might not conform to some domain-specific constraints, and (3) inaccuracy: due to the inherent uncertainty in LLMs, the models can include inaccurate, hallucinated elements. While the first issue is often addressed through techniques such as constraint decoding or filtering, the latter two remain largely unaddressed. Motivated by recent self-consistency approaches in LLMs, we propose a novel abstraction-concretization framework that enhances the consistency and quality of generated graph models by considering multiple outputs from an LLM. Our approach first constructs a probabilistic partial model that aggregates all candidate outputs and then refines this partial model into the most appropriate concrete model that satisfies all constraints. We evaluate our framework on several popular open-source and closed-source LLMs using diverse datasets for model generation tasks. The results demonstrate that our approach significantly improves both the consistency and quality of the generated graph models.

Medical vision foundational models are used for a wide variety of tasks, including medical image segmentation and registration. This work evaluates the ability of these models to predict disease progression using a simple linear probe. We hypothesize that intermediate layer features of segmentation models capture structural information, while those of registration models encode knowledge of change over time. Beyond demonstrating that these features are useful for disease progression prediction, we also show that registration model features do not require spatially aligned input images. However, for segmentation models, spatial alignment is essential for optimal performance. Our findings highlight the importance of spatial alignment and the utility of foundation model features for image registration.

Requirements over strings, commonly represented using natural language (NL), are particularly relevant for software systems due to their heavy reliance on string data manipulation. While individual requirements can usually be analyzed manually, verifying properties (e.g., satisfiability) over sets of NL requirements is particularly challenging. Formal approaches (e.g., SMT solvers) may efficiently verify such properties, but are known to have theoretical limitations. Additionally, the translation of NL requirements into formal constraints typically requires significant manual effort. Recently, large language models (LLMs) have emerged as an alternative approach for formal reasoning tasks, but their effectiveness in verifying requirements over strings is less studied. In this paper, we introduce a hybrid approach that verifies the satisfiability of NL requirements over strings by using LLMs (1) to derive a satisfiability outcome (and a consistent string, if possible), and (2) to generate declarative (i.e., SMT) and imperative (i.e., Python) checkers, used to validate the correctness of (1). In our experiments, we assess the performance of four LLMs. Results show that LLMs effectively translate natural language into checkers, even achieving perfect testing accuracy for Python-based checkers. These checkers substantially help LLMs in generating a consistent string and accurately identifying unsatisfiable requirements, leading to more than doubled generation success rate and F1-score in certain cases compared to baselines without generated checkers.

Language models of code have demonstrated remarkable performance across various software engineering and source code analysis tasks. However, their demanding computational resource requirements and consequential environmental footprint remain as significant challenges. This work introduces Alpine , an adaptive programming language-agnostic pruning technique designed to substantially reduce the computational overhead of these models. The proposed method offers a pluggable layer that can be integrated with all Transformer-based models. With Alpine , input sequences undergo adaptive compression throughout the pipeline, reaching a size that is up to ×3 less their initial size, resulting in significantly reduced computational load. Our experiments on two software engineering tasks, defect prediction and code clone detection across three language models CodeBert , GraphCodeBert and UniXCoder show that Alpine achieves up to a 50% reduction in FLOPs, a 58.1% decrease in memory footprint, and a 28.1% improvement in throughput on average. This led to a reduction in CO 2 emissions by up to 44.85%. Importantly, it achieves a reduction in computation resources while maintaining up to 98.1% of the original predictive performance. These findings highlight the potential of Alpine in making language models of code more resource-efficient and accessible while preserving their performance, contributing to the overall sustainability of their adoption in software development. Also, it sheds light on redundant and noisy information in source code analysis corpora, as shown by the substantial sequence compression achieved by Alpine .

Motivation: Automated bug detection in dynamically typed languages such as Python is essential for maintaining code quality. The lack of mandatory type annotations in such languages can lead to errors that are challenging to identify early with traditional static analysis tools. Recent progress in deep neural networks has led to increased use of neural bug detectors. In statically typed languages, a type checker is integrated into the compiler and thus taken into consideration when the neural bug detector is designed for these languages. Problem: However, prior studies overlook this aspect during the training and testing of neural bug detectors for dynamically typed languages. When an optional type checker is used, assessing existing neural bug detectors on bugs easily detectable by type checkers may impact their performance estimation. Moreover, including these bugs in the training set of neural bug detectors can shift their detection focus toward the wrong type of bugs. Contribution: We explore the impact of type checking on various neural bug detectors for variable misuse bugs, a common type targeted by neural bug detectors. Existing synthetic and real-world datasets are type-checked to evaluate the prevalence of type-related bugs. Then, we investigate how type-related bugs influence the training and testing of the neural bug detectors. Findings: Our findings indicate that existing bug detection datasets contain a significant proportion of type-related bugs. Building on this insight, we discover integrating the neural bug detector with a type checker can be beneficial, especially when the code is annotated with types. Further investigation reveals neural bug detectors perform better on type-related bugs than other bugs. Moreover, removing type-related bugs from the training data helps improve neural bug detectors' ability to identify bugs beyond the scope of type checkers.

Recently, large language models (LLMs) have achieved widespread application across various fields. Despite their impressive capabilities, LLMs suffer from a lack of structured reasoning ability, particularly for complex tasks requiring domain-specific best practices, which are often unavailable in the training data. Although multi-step prompting methods incorporating human best practices, such as chain-of-thought and tree-of-thought, have gained popularity, they lack a general mechanism to control LLM behavior. In this paper, we propose SHERPA, a model-driven framework to improve the LLM performance on complex tasks by explicitly incorporating domain-specific best practices into hierarchical state machines. By structuring the LLM execution processes using state machines, SHERPA enables more fine-grained control over their behavior via rules or decisions driven by machine learning-based approaches, including LLMs. We show that SHERPA is applicable to a wide variety of tasks-specifically, code generation, class name generation, and question answering-replicating previously proposed approaches while further improving the performance. We demonstrate the effectiveness of SHERPA for the aforementioned tasks using various LLMs. Our systematic evaluation compares different state machine configurations against baseline approaches without state machines. Results show that integrating well-designed state machines significantly improves the quality of LLM outputs, and is particularly beneficial for complex tasks with well-established human best practices but lacking data used for training LLMs.

In many critical domains, features are not freely available at inference time: each measurement may come with a cost of time, money, and risk. Longitudinal prediction further complicates this setting because both features and labels evolve over time, and missing measurements at earlier timepoints may become permanently unavailable. We propose NOCTA, a Non-Greedy Objective Cost-Tradeoff Acquisition framework that sequentially acquires the most informative features at inference time while accounting for both temporal dynamics and acquisition cost. NOCTA is driven by a novel objective, NOCT, which evaluates a candidate set of future feature-time acquisitions by its expected predictive loss together with its acquisition cost. Since NOCT depends on unobserved future trajectories at inference time, we develop two complementary estimators: (i) NOCT-Contrastive, which learns an embedding of partial observations utilizing the induced distribution over future acquisitions, and (ii) NOCT-Amortized, which directly predicts NOCT for candidate plans with a neural network. Experiments on synthetic and real-world medical datasets demonstrate that both NOCTA estimators outperform existing baselines, achieving higher accuracy at lower acquisition costs.

Behavioral model diagrams, e.g., sequence diagrams, are an essential form of documentation that are typically designed by system engineers from requirements documentation, either fully manually or assisted by design tools. With the growing use of Large Language Models (LLM) as AI modeling assistants, more automation will be involved in generating diagrams. This necessitates the advancement of automatic model correctness evaluation tools. Such a tool can be used to evaluate both manually and AI automatically generated models; to provide feedback to system engineers, and enable AI assistants to self-evaluate and self-enhance their generated models. In this paper, we propose MCeT, the first fully automated tool to evaluate the correctness of a behavioral model, sequence diagrams in particular, against its corresponding requirements text and produce a list of issues that the model has. We utilize LLMs for the correctness evaluation tasks as they have shown outstanding natural language understanding ability. However, we show that directly asking an LLM to compare a diagram to requirements finds less than 35% of issues that experienced engineers can find. We propose to supplement the direct check with a fine-grained, multi-perspective approach; we split the diagram into atomic, non-divisible interactions, and split the requirements text into atomic, self-contained items. We compare the diagram with atomic requirements and each diagramatom with the requirements. We also propose a self-consistency checking approach that combines perspectives to mitigate LLM hallucinated issues. Our combined approach improves upon the precision of the direct approach from 0.58 to 0.81 in a dataset of real requirements. Moreover, the approach finds 90% more issues that the experienced engineers found than the direct approach, and reports an average of 6 new issues per diagram.

Motivation . Large language models (LLMs) have exhibited remarkable proficiency in diverse software engineering (SE) tasks, such as code summarization, code translation, and code search. Handling such tasks typically involves acquiring foundational coding knowledge on large, general-purpose datasets during a pre-training phase, and subsequently refining on smaller, task-specific datasets as part of a fine-tuning phase. Problem statement. Data leakage i.e., using information of the test set to perform the model training, is a well-known issue in training of machine learning models. A manifestation of this issue is the intersection of the training and testing splits. While intra-dataset code duplication examines this intersection within a given dataset and has been addressed in prior research, inter-dataset code duplication, which gauges the overlap between different datasets, remains largely unexplored. If this phenomenon exists, it could compromise the integrity of LLM evaluations because of the inclusion of fine-tuning test samples that were already encountered during pre-training, resulting in inflated performance metrics. Contribution. This paper explores the phenomenon of inter-dataset code duplication and its impact on evaluating LLMs across diverse SE tasks. Study design. We conduct an empirical study using the CodeSearchNet dataset (CSN), a widely adopted pre-training dataset, and five fine-tuning datasets used for various SE tasks. We first identify the intersection between the pre-training and fine-tuning datasets using a deduplication process. Next, we pre-train two versions of LLMs using a subset of CSN: one leaky LLM, which includes the identified intersection in its pre-training set, and one non-leaky LLM that excludes these samples. Finally, we fine-tune both models and compare their performances using fine-tuning test samples that are part of the intersection. Results. Our findings reveal a potential threat to the evaluation of LLMs across multiple SE tasks, stemming from the inter-dataset code duplication phenomenon. We also demonstrate that this threat is accentuated by the chosen fine-tuning technique. Furthermore, we provide evidence that open-source models such as CodeBERT, GraphCodeBERT, and UnixCoder could be affected by inter-dataset duplication. Based on our findings, we delve into prior research that may be susceptible to this threat. Additionally, we offer guidance to SE researchers on strategies to prevent inter-dataset code duplication.

Large language models (LLMs) are being increasingly adopted in the software engineering domain, yet the robustness of their grasp on core software design concepts remains unclear. We conduct an empirical study to systematically evaluate their understanding of cohesion (intra-module) and coupling (inter-module). We programmatically generate poorly designed code fragments and test the DeepSeek-R1 model family (14B, 32B, 70B) under varying levels of guidance, from simple Verification to Guided and Open-ended Generation, while varying contextual noise by injecting distractor elements. While models exhibit a solid baseline understanding of both concepts in ideal conditions, their practical knowledge is fragile and highly asymmetrical. Reasoning about coupling proves brittle; performance collapses in noisy, open-ended scenarios, with F1 scores dropping by over 50%. In contrast, the models' analysis of cohesion is remarkably robust to internal noise in guided tasks, showing little performance degradation. However, this resilience also fails when all guidance is removed. Reasoning-trace analysis confirms these failure modes, revealing cognitive shortcutting for coupling versus a more exhaustive (yet still failing) analysis for cohesion. To summarize, while LLMs can provide reliable assistance for recognizing design flaws, their ability to reason autonomously in noisy, realistic contexts is limited, highlighting the critical need for more scalable and robust program understanding capabilities.

Welcome to the 15th International Workshop on Model-Driven Requirements Engineering (MoDRE ' 25), held in conjunction with the 33rd edition of the Requirements Engineering Conference. The MoDRE workshop series has established a forum where researchers and practitioners can discuss the challenges and opportunities of Model-Driven Development (MDD) for Requirements Engineering (RE). Model-driven (software) development languages, tools, and techniques have raised the level of abstraction in software development and enabled automation of various parts of the software development process. When effectively applied, MDD techniques can offer significant benefits to RE, by balancing the flexibility for capturing varied user needs with the formality required for model transformations, and by bridging high-level abstraction with the richness of requirements information. MoDRE seeks to explore areas of RE that are not yet fully formalized to be incorporated into an MDD environment. It also seeks to explore how RE models can benefit from advances in the model-driven community, such as flexible, collaborative, and AI-enabled modeling. MoDRE encourages researchers to explore these benefits by identifying new challenges, sharing ongoing work and emerging solutions, analyzing strengths and weaknesses of MDD approaches for RE, and fostering stimulating discussions on the topic during the workshop. This workshop is an opportunity to reflect on the current state and envision the future of MDD approaches for RE. We would like to thank the Program Committee for their valuable feedback to the authors, and, of course, the authors for submitting their papers and making this workshop possible.

Large Language Models have significantly advanced the field of code generation, demonstrating the ability to produce functionally correct code snippets. However, advancements in generative AI for code overlook foundational Software Engineering (SE) principles such as modularity, and single responsibility, and concepts such as cohesion and coupling which are critical for creating maintainable, scalable, and robust software systems. These concepts are missing in pipelines that start with pre-training and end with the evaluation using benchmarks. This vision paper argues for the integration of SE knowledge into LLMs to enhance their capability to understand, analyze, and generate code and other SE artifacts following established SE knowledge. The aim is to propose a new direction where LLMs can move beyond mere functional accuracy to perform generative tasks that require adherence to SE principles and best practices. In addition, given the interactive nature of these conversational models, we propose using Bloom's Taxonomy as a framework to assess the extent to which they internalize SE knowledge. The proposed evaluation framework offers a sound and more comprehensive evaluation technique compared to existing approaches such as linear probing. Software engineering native generative models will not only overcome the shortcomings present in current models but also pave the way for the next generation of generative models capable of handling real-world software engineering.

Graph convolutional neural networks (GCNs) are powerful tools for learning graph-based knowledge representations from training data. However, they are vulnerable to small perturbations in the input graph, which makes them susceptible to input faults or adversarial attacks. This poses a significant problem for GCNs intended to be used in critical applications, which need to provide certifiably robust services even in the presence of adversarial perturbations. We propose an improved GCN robustness certification technique for node classification in the presence of node feature perturbations. We introduce a novel polyhedra-based abstract interpretation approach to tackle specific challenges of graph data and provide tight upper and lower bounds for the robustness of the GCN. Experiments show that our approach simultaneously improves the tightness of robustness bounds as well as the runtime performance of certification. Moreover, our method can be used during training to further improve the robustness of GCNs.

Behavior-driven development (BDD) enables collaboration among different stakeholders by employing a natural-language representation of system requirements and of test scenarios. These scenarios often involve constraints over string values, e.g. for the validity of email addresses, which are challenging to test comprehensively. Traditional methods like SMT solvers (e.g. Z3, Ostrich) handle constraints efficiently but produce unrealistic strings and require formal specifications that are often unavailable and expensive to compute. This paper explores the potential of large language models (LLMs) in generating realistic, constraint-satisfying strings for BDD. We propose an evaluation framework to assess LLMs’ ability to (1) generate consistent string values and (2) detect constraint inconsistencies. In our experiments, three LLMs are compared to state-of-the-art solvers using constraints from a software engineering course project. Results show that while solvers dominate in precision and recall, LLMs derive realistic strings more suitable for a requirements engineering context. With these trade-offs, we believe that, when formal constraints are available, a combined LLM-solver approach could offer a more effective solution.

Domain modeling is an essential component in many software engineering courses since it serves as a way to represent and understand the concepts and relationships in a problem domain. Course instructors evaluate student-generated diagrams manually, comparing them against a reference solution and providing feedback. However, as enrollment in software engineering courses continues to rise, manual grading of a large number of student submissions becomes an overwhelming and time-intensive task for instructors. Hence, there is a need for automated assessment of domain models which assists course instructors during the grading process. In this paper, we propose a novel text embedding-based approach that automatizes the assessment of domain models expressed in a textual domain-specific language, against reference solutions created by modeling experts. Our algorithm showcases remarkable proficiency in matching model elements across domain models, achieving an F1-score of 0.82 for class matching, 0.75 for attribute matching, and 0.80 for relation matching. Our algorithm also yields grades highly correlated with human grader assessments, with correlations exceeding 0.8 and mean absolute errors below 0.05.