arXiv subjects

Oussama Ben Sghaier

Publications and source records attributed to Oussama Ben Sghaier.

12 recordsLinked to original sources

arXiv2026

When Model Release Meets Model Reuse: Producer-Consumer Misalignment in Hugging Face

Pre-trained Language Models (PTLMs) are increasingly reused as dependencies in modern software systems, even though prior work has documented persistent structural problems in PTLM supply chains, such as inconsistent release practices, incomplete metadata, and divergence between Hugging Face and GitHub repositories. One previously unexplored angle on these problems is that misalignment between PTLM producers' release practices and consumers' model reuse needs may explain several of these challenges, yet the human expectations, processes and interactions underneath this misalignment have never been studied. As such, we conducted the first dual-perspective survey of 50 Hugging Face producers and 95 GitHub consumers to examine producer-consumer misalignments across four AI supply chain dimensions: model discovery, documentation practices, lineage tracing, and model governance adoption. We find that, although producers and consumers rely on the same documentation artifacts, they disagree on where critical metadata should be recorded. Even though 27.9% of producers and 31.4% of consumers trace lineage beyond the immediate parent model, producers primarily trace model lineage for provenance and reproducibility, whereas consumers are mainly driven by quality and reliability concerns. Furthermore, producers believe that governance mechanisms that streamline model release would have the greatest positive impact, whereas consumers prioritize documentation and dependency transparency. These findings highlight opportunities to improve model documentation conventions, lineage visibility, and governance support in PTLM supply chains.

cs.SE↗

arXiv2026

Don't Blame the Large Language Model: How Agent Harness Evolution Shapes Coding Agent Quality

Coding agents, autonomous systems that use large language models (LLMs) to resolve software engineering tasks, rely on agent harness: a middleware layer in between a developer and a large language model that orchestrates system prompts, tool execution, context management, and iterative reasoning loops. While these agent harnesses evolve at extreme velocities, no study has examined how this evolution affects agent quality (i.e., effectiveness and efficiency) over time. Practitioners regularly report quality regressions after agent harness updates, yet consistently attribute them to the underlying model rather than the harness itself. In this paper, we address this gap by conducting the first controlled longitudinal study that isolates the agent harness contribution. Unlike prior work that fixes the agent harness and varies the model, we fix the model and vary only the agent harness, evaluating 35 sequential releases to measure their impact on agent effectiveness and efficiency. We first empirically study the development and release evolution of five major open-source agent harnesses (i.e., Codex, Qwen Code, Gemini, OpenCode, and OpenHands), revealing extreme release velocities exceeding two releases per day and thousands of issues within months. We then perform a controlled deep dive into 35 sequential releases of the Qwen Code CLI, evaluating each against 50 stratified SWE-bench Verified tasks while holding the underlying LLM constant. We trace the resulting quality fluctuations to specific development patterns and architectural components, and illustrate our findings with concrete qualitative evidence linking individual pull requests to measured quality shifts.

cs.SE↗

arXiv2026

Balancing Usefulness and Naturalness: An LLM-based Curation Pipeline for Code Review Comments

Code review is a cornerstone of software development, where reviewers provide feedback through written comments to ensure code quality, maintainability, and correctness. The effectiveness of this process hinges on the quality of review comments. As large language models (LLMs) gain traction in automating code review tasks, the utility of these systems is directly limited by the quality of the datasets on which they are trained. Unfortunately, existing code review datasets are often noisy, inconsistent, or poorly structured, which hinders the ability of LLMs to learn to generate accurate, helpful, and human-like review comments. To overcome these limitations, we propose two different curation pipelines designed to improve both the quality and the utility of large-scale code review datasets. In the first pipeline, all review comments are systematically reformulated by an LLM to improve their clarity, conciseness, and civility while preserving their semantic intent. The curated dataset resulting from this approach, called CuREV, offers cleaner, higher-quality, and easier-to-learn-from comments that lead to measurable improvements in downstream automation tasks, namely review comment generation and code refinement. Building on this, we propose an improved pipeline, guided by high-quality exemplars, that enhances the realism and diversity of curated review comments. This method first separates the dataset into high-quality and low-quality reviews, based on a systematic quality assessment using an evaluation framework. High-quality comments are preserved in their original form and further used as in-context exemplars to inspire the reformulation of low-quality comments. By varying the exemplars provided, the reformulated comments are not only clearer and more actionable but also exhibit a broader range of writing styles, making them more realistic and human-like.

cs.SE↗

arXiv2026

Towards Imputation of Pre-Trained Language Model Metadata using Semantic Fingerprinting

Pre-trained language models (PTLMs) hosted on platforms such as Hugging Face form complex lineage structures similar to software dependency graphs. However, unlike traditional software ecosystems, PTLM repositories often lack reliable provenance due to missing metadata, such as licenses, reuse methods, pipeline tags, model types, and training libraries. To address this gap, we introduce Semantic Fingerprinting (SemFin), a lightweight approach that combines Hugging Face (HF) configuration files with model repository tags to automatically impute missing model metadata fields and reconstruct model lineage chains. We evaluate SemFin on a large-scale dataset of 317,133 PTLMs. Our results show that configuration files typically encode the technical requirements necessary to instantiate and reuse models, enabling them to serve as a structural blueprint for model reuse, particularly for transformer-based architectures. By combining these configuration files with model repository tags, SemFin significantly outperforms the existing propagation-based imputation approaches, improving prediction accuracy by up to 31.4% and 26.6% compared to Graph Avg and Hub Avg baselines. Importantly, SemFin also imputes metadata for 16.6% of isolated models where propagation-based methods fail. Applying SemFin to impute missing reuse-method and license metadata for 167,089 unlabeled models reveals that traceable reuse method chains expand by 75.9% and license lineage chains by 53.6%, uncovering 86 previously invisible reuse method patterns, while the proportion of incompatible license patterns only increases from 34.8% to 36.8%. These findings demonstrate how automatically derived structural signals can support the automated construction of AI Bills of Materials (AIBOMs), helping transform metadata from an error-prone manual declaration into information inferred directly from model artifacts.

cs.SE↗

arXiv2026

On the synchronization between Hugging Face pre-trained language models and their upstream GitHub repository

Pre-trained language models (PTLMs) have transformed natural language processing (NLP), enabling major advances in tasks such as text generation and translation. Similar to software package management, PTLMs are developed using code and environment scripts hosted in upstream repositories (e.g., GitHub), while families of trained model variants are distributed through downstream platforms such as Hugging Face (HF). Despite this similarity, coordinating development and release activities across these platforms remains challenging, leading to misaligned timelines, inconsistent versioning practices, and barriers to effective reuse. To examine how commit activities are coordinated between GitHub and HF, we conducted an in-depth mixed-method study of 325 PTLM families comprising 904 HF model variants. Our findings show that GitHub contributors primarily focus on model version specification, code quality improvements, performance optimization, and dependency management, whereas HF contributors mainly address model documentation, dataset handling, and inference setup. We further analyze synchronization across three dimensions -- lag, type, and intensity -- revealing eight distinct synchronization patterns. The dominance of partially synchronized patterns, such as Disperse and Sparse synchronization, highlights structural disconnects in cross-platform release practices. These disconnects often result in isolated or abandoned updates, increasing the risk of incomplete, outdated, or behaviorally inconsistent models being exposed to end users. Recognizing these synchronization patterns is essential for improving oversight and traceability in PTLM release workflows.

cs.SE↗

arXiv2025

Artificial or Just Artful? Do LLMs Bend the Rules in Programming?

Large Language Models (LLMs) are widely used for automated code generation, yet their apparent successes often mask a tension between pretraining objectives and alignment choices. While pretraining encourages models to exploit all available signals to maximize success, alignment, whether through fine-tuning or prompting, may restrict their use. This conflict is especially salient in agentic AI settings, for instance when an agent has access to unit tests that, although intended for validation, act as strong contextual signals that can be leveraged regardless of explicit prohibitions. In this paper, we investigate how LLMs adapt their code generation strategies when exposed to test cases under different prompting conditions. Using the BigCodeBench (Hard) dataset, we design five prompting conditions that manipulate test visibility and impose explicit or implicit restrictions on their use. We evaluate five LLMs (four open-source and one closed-source) across correctness, code similarity, program size, and code churn, and analyze cross-model consistency to identify recurring adaptation strategies. Our results show that test visibility dramatically alters performance, correctness nearly doubles for some models, while explicit restrictions or partial exposure only partially mitigate this effect. Beyond raw performance, we identify four recurring adaptation strategies, with test-driven refinement emerging as the most frequent. These results highlight how LLMs adapt their behavior when exposed to contextual signals that conflict with explicit instructions, providing useful insight into how models reconcile pretraining objectives with alignment constraints.

cs.SE↗

arXiv2025

Leveraging Reward Models for Guiding Code Review Comment Generation

Code review is a crucial component of modern software development, involving the evaluation of code quality, providing feedback on potential issues, and refining the code to address identified problems. Despite these benefits, code review can be rather time consuming, and influenced by subjectivity and human factors. For these reasons, techniques to (partially) automate the code review process have been proposed in the literature. Among those, the ones exploiting deep learning (DL) are able to tackle the generative aspect of code review, by commenting on a given code as a human reviewer would do (i.e., comment generation task) or by automatically implementing code changes required to address a reviewer's comment (i.e., code refinement task). In this paper, we introduce CoRAL, a deep learning framework automating review comment generation by exploiting reinforcement learning with a reward mechanism considering both the semantics of the generated comments as well as their usefulness as input for other models automating the code refinement task. The core idea is that if the DL model generates comments that are semantically similar to the expected ones or can be successfully implemented by a second model specialized in code refinement, these comments are likely to be meaningful and useful, thus deserving a high reward in the reinforcement learning framework. We present both quantitative and qualitative comparisons between the comments generated by CoRAL and those produced by the latest baseline techniques, highlighting the effectiveness and superiority of our approach.

cs.SE↗

arXiv2025

Combining Large Language Models with Static Analyzers for Code Review Generation

Code review is a crucial but often complex, subjective, and time-consuming activity in software development. Over the past decades, significant efforts have been made to automate this process. Early approaches focused on knowledge-based systems (KBS) that apply rule-based mechanisms to detect code issues, providing precise feedback but struggling with complex, context-dependent cases. More recent work has shifted toward fine-tuning pre-trained language models for code review, enabling broader issue coverage but often at the expense of precision. In this paper, we propose a hybrid approach that combines the strengths of KBS and learning-based systems (LBS) to generate high-quality, comprehensive code reviews. Our method integrates knowledge at three distinct stages of the language model pipeline: during data preparation (Data-Augmented Training, DAT), at inference (Retrieval-Augmented Generation, RAG), and after inference (Naive Concatenation of Outputs, NCO). We empirically evaluate our combination strategies against standalone KBS and LBS fine-tuned on a real-world dataset. Our results show that these hybrid strategies enhance the relevance, completeness, and overall quality of review comments, effectively bridging the gap between rule-based tools and deep learning models.

cs.SE↗

arXiv2025

Harnessing Large Language Models for Curated Code Reviews

In code review, generating structured and relevant comments is crucial for identifying code issues and facilitating accurate code changes that ensure an efficient code review process. Well-crafted comments not only streamline the code review itself but are also essential for subsequent tasks like code refinement, where the code is modified to satisfy the input review comment. Although various AI-based approaches aimed to automate comment generation, their effectiveness remains limited by the quality of the training data. Existing code review datasets are often noisy and unrefined, posing limitations to the learning potential of AI models and hindering the automation process. To address these challenges, we propose a curation pipeline designed to enhance the quality of the largest publicly available code review dataset. We begin by establishing an evaluation framework, incorporating specific criteria and categories to empirically study the initial quality of the dataset. Using a large language model (LLM)-driven approach, we then apply our curation pipeline to refine the dataset. A comparative analysis of the newly curated dataset, based on the same evaluation framework, demonstrates substantial improvements in the clarity and conciseness of the comments. Additionally, we assess the impact of the curated dataset on automating downstream tasks, specifically comment generation and code refinement. Our findings show that the curated dataset leads to enhanced model performance in generating more accurate comments. Curated comments are also more useful as they lead to more accurate code refinement.

cs.SE↗

arXiv2024

Improving the Learning of Code Review Successive Tasks with Cross-Task Knowledge Distillation

Code review is a fundamental process in software development that plays a pivotal role in ensuring code quality and reducing the likelihood of errors and bugs. However, code review can be complex, subjective, and time-consuming. Quality estimation, comment generation, and code refinement constitute the three key tasks of this process, and their automation has traditionally been addressed separately in the literature using different approaches. In particular, recent efforts have focused on fine-tuning pre-trained language models to aid in code review tasks, with each task being considered in isolation. We believe that these tasks are interconnected, and their fine-tuning should consider this interconnection. In this paper, we introduce a novel deep-learning architecture, named DISCOREV, which employs cross-task knowledge distillation to address these tasks simultaneously. In our approach, we utilize a cascade of models to enhance both comment generation and code refinement models. The fine-tuning of the comment generation model is guided by the code refinement model, while the fine-tuning of the code refinement model is guided by the quality estimation model. We implement this guidance using two strategies: a feedback-based learning objective and an embedding alignment objective. We evaluate DISCOREV by comparing it to state-of-the-art methods based on independent training and fine-tuning. Our results show that our approach generates better review comments, as measured by the BLEU score, as well as more accurate code refinement according to the CodeBLEU score

cs.SE↗

arXiv2023

Toward Optimal Psychological Functioning in AI-driven Software Engineering Tasks: The SEWELL-CARE Assessment Framework

In the field of software engineering, there has been a shift towards utilizing various artificial intelligence techniques to address challenges and create innovative tools. These solutions are aimed at enhancing efficiency, automating tasks, and providing valuable support to developers. While the technical aspects are crucial, the well-being and psychology of the individuals performing these tasks are often overlooked. This paper argues that a holistic approach is essential, one that considers the technical, psychological, and social aspects of software engineering tasks. To address this gap, we introduce SEWELL-CARE, a conceptual framework designed to assess AI-driven software engineering tasks from multiple perspectives, with the goal of customizing the tools to improve the efficiency, well-being, and psychological functioning of developers. By emphasizing both technical and human dimensions, our framework provides a nuanced evaluation that goes beyond traditional technical metrics.

cs.SE↗

arXiv2023

Unity is Strength: Cross-Task Knowledge Distillation to Improve Code Review Generation

Code review is a fundamental process in software development that plays a critical role in ensuring code quality and reducing the likelihood of errors and bugs. However, code review might be complex, subjective, and time-consuming. Comment generation and code refinement are two key tasks of this process and their automation has traditionally been addressed separately in the literature using different approaches. In this paper, we propose a novel deep-learning architecture, DISCOREV, based on cross-task knowledge distillation that addresses these two tasks simultaneously. In our approach, the fine-tuning of the comment generation model is guided by the code refinement model. We implemented this guidance using two strategies, feedback-based learning objective and embedding alignment objective. We evaluated our approach based on cross-task knowledge distillation by comparing it to the state-of-the-art methods that are based on independent training and fine-tuning. Our results show that our approach generates better review comments as measured by the BLEU score.

cs.SE↗