arXiv subjects

Bram Adams

Publications and source records attributed to Bram Adams.

At least 37 records · Page 2Linked to original sources

arXiv2026

Model Context Protocol (MCP) at First Glance: Studying the Security and Maintainability of MCP Servers

Although Foundation Models (FMs), such as GPT-4, are increasingly used in domains like finance and software engineering, reliance on textual interfaces limits these models' real-world interaction. To address this, FM providers introduced a tool called -- triggering a proliferation of frameworks with distinct tool interfaces. In late 2024, Anthropic introduced the Model Context Protocol (MCP) to standardize this tool ecosystem. MCP is rapidly emerging as a de facto industry standard. Despite its adoption, MCP's AI-driven, non-deterministic control flow introduces new risks to sustainability, security, and maintainability, warranting closer examination. Towards this end, we present the first large-scale empirical study of MCP. Using state-of-the-art health metrics and a hybrid analysis pipeline that combines a general-purpose static analysis tool with an MCP-specific scanner, we evaluate 1,899 open-source MCP servers to assess their health, security, and maintainability. Despite MCP servers demonstrating strong health metrics, we identify eight distinct vulnerabilities -- only three of which overlap with traditional software vulnerabilities. Additionally, 7.2% of servers contain general vulnerabilities, and 5.5% exhibit MCP-specific tool poisoning. Regarding maintainability, while 66% exhibit code smells, 14.4% contain ten bug patterns overlapping prior research. These findings highlight the need for MCP-specific vulnerability detection techniques while reaffirming the value of traditional analysis and refactoring practices. Furthermore, we advocate for stronger governance across the MCP ecosystem by incorporating MCP-specific vulnerabilities into standardized vulnerability databases, enabling automated security scanning within MCP registries, and promoting responsible development practices to ensure the long-term safety and sustainability of the MCP ecosystem.

cs.SE↗

arXiv2026

An Empirical Study of Testing Practices in Open Source AI Agent Frameworks and Agentic Applications

Foundation model (FM)-based AI agents are rapidly gaining adoption across diverse domains, but their inherent non-determinism and non-reproducibility pose testing and quality assurance challenges. While recent benchmarks provide task-level evaluations, there is limited understanding of how developers verify the internal correctness of these agents during development. To address this gap, we conduct the first large-scale empirical study of testing practices in the AI agent ecosystem, analyzing 39 open-source agent frameworks and 439 agentic applications. We identify ten distinct testing patterns and find that novel, agent-specific methods like DeepEval are seldom used (around 1%), while traditional patterns like negative and membership testing are widely adapted to manage FM uncertainty. By mapping these patterns to canonical architectural components of agent frameworks and agentic applications, we uncover a fundamental inversion of testing effort: deterministic components like Resource Artifacts (tools) and Coordination Artifacts (workflows) consume over 70% of testing effort, while the FM-based Plan Body receives less than 5%. Crucially, this reveals a critical blind spot, as the Trigger component (prompts) remains neglected, appearing in around 1% of all tests. Our findings offer the first empirical testing baseline in FM-based agent frameworks and agentic applications, revealing a rational but incomplete adaptation to non-determinism. To address it, framework developers should improve support for novel testing methods, application developers must adopt prompt regression testing, and researchers should explore barriers to adoption. Strengthening these practices is vital for building more robust and dependable AI agents.

cs.SE↗

arXiv2026

HAFixAgent: History-Aware Program Repair Agent

Automated program repair (APR) has recently shifted toward large language models and agent-based systems, yet most systems rely on local snapshot context, overlooking repository history. Prior work shows that repository history helps repair single-line bugs, since the last commit touching the buggy line is often the bug-introducing one. In this paper, we investigate whether repository history can also improve agentic APR systems at scale, especially for complex multi-hunk bugs. We present HAFixAgent, a History-Aware Bug-Fixing Agent that injects blame-derived repository heuristics into its repair loop. A preliminary study on 854 Defects4J (Java) and 501 BugsInPy (Python) bugs motivates our design, showing that bug-relevant history is widely available across both benchmarks. Using the same LLM (DeepSeek-V3.2-Exp) for all experiments, including replicated baselines, we show: (1) Effectiveness: HAFixAgent outperforms RepairAgent (+56.6\%) and BIRCH-feedback (+47.1\%) on Defects4J. Historical context further improves repair by +4.4\% on Defects4J and +38.6\% on BugsInPy, especially on single-file multi-hunk (SFMH) bugs. (2) Robustness: under noisy fault localization (+1/+3/+5 line shifts), history provides increasing resilience, maintaining 40 to 56\% success on SFMH bugs where the non-history baseline collapses to 0\%. (3) Efficiency: history does not significantly increase agent steps or token costs on either benchmark.

cs.SE↗

arXiv2026

Human-AI Synergy in Agentic Code Review

Code review is a critical software engineering practice where developers review code changes before integration to ensure code quality, detect defects, and improve maintainability. In recent years, AI agents that can understand code context, plan review actions, and interact with development environments have been increasingly integrated into the code review process. However, there is limited empirical evidence to compare the effectiveness of AI agents and human reviewers in collaborative workflows. To address this gap, we conduct a large-scale empirical analysis of 278,790 code review conversations across 300 open-source GitHub projects. In our study, we aim to compare the feedback differences provided by human reviewers and AI agents. We investigate human-AI collaboration patterns in review conversations to understand how interaction shapes review outcomes. Moreover, we analyze the adoption of code suggestions provided by human reviewers and AI agents into the codebase and how adopted suggestions change code quality. We find that human reviewers provide additional feedback than AI agents, including understanding, testing, and knowledge transfer. Human reviewers exchange 11.8% more rounds when reviewing AI-generated code than human-written code. Moreover, code suggestions made by AI agents are adopted into the codebase at a significantly lower rate than suggestions proposed by human reviewers. Over half of unadopted suggestions from AI agents are either incorrect or addressed through alternative fixes by developers. When adopted, suggestions provided by AI agents produce significantly larger increases in code complexity and code size than suggestions provided by human reviewers. Our findings suggest that while AI agents can scale defect screening, human oversight remains critical for ensuring suggestion quality and providing contextual feedback that AI agents lack.

cs.SE↗

arXiv2026

From Leaderboard to Deployment: Code Quality Challenges in AV Perception Repositories

Autonomous vehicle (AV) perception models are typically evaluated solely on benchmark performance metrics, with limited attention to code quality, production readiness and long-term maintainability. This creates a significant gap between research excellence and real-world deployment in safety-critical systems subject to international safety standards. To address this gap, we present the first large-scale empirical study of software quality in AV perception repositories, systematically analyzing 178 unique models from the KITTI and NuScenes 3D Object Detection leaderboards. Using static analysis tools (Pylint, Bandit, and Radon), we evaluated code errors, security vulnerabilities, maintainability, and development practices. Our findings revealed that only 7.3% of the studied repositories meet basic production-readiness criteria, defined as having zero critical errors and no high-severity security vulnerabilities. Security issues are highly concentrated, with the top five issues responsible for almost 80% of occurrences, which prompted us to develop a set of actionable guidelines to prevent them. Additionally, the adoption of Continuous Integration/Continuous Deployment pipelines was correlated with better code maintainability. Our findings highlight that leaderboard performance does not reflect production readiness and that targeted interventions could substantially improve the quality and safety of AV perception code.

cs.CV↗

arXiv2026

HAFix: History-Augmented Large Language Models for Bug Fixing

Recent studies have explored the performance of Large Language Models (LLMs) on various Software Engineering (SE) tasks, such as code generation and bug fixing. However, these approaches typically rely on the context data from the current snapshot of the project, overlooking the potential of rich historical data residing in real-world software repositories. Additionally, the impact of prompt styles on LLM performance for SE tasks within a historical context remains underexplored. To address these gaps, we propose HAFix, which stands for History-Augmented LLMs on Bug Fixing, a novel approach that leverages seven individual historical heuristics associated with bugs and aggregates the results of these heuristics (HAFix-Agg) to enhance LLMs' bug-fixing capabilities. To empirically evaluate HAFix, we employ three Code LLMs (i.e., Code Llama, DeepSeek-Coder and DeepSeek-Coder-V2-Lite models) on 51 single-line Python bugs from BugsInPy and 116 single-line Java bugs from Defects4J. Our evaluation demonstrates that multiple HAFix heuristics achieve statistically significant improvements compared to a non-historical baseline inspired by GitHub Copilot. Furthermore, the aggregated HAFix variant HAFix-Agg achieves substantial improvements by combining the complementary strengths of individual heuristics, increasing bug-fixing rates by an average of 45.05% on BugsInPy and 49.92% on Defects4J relative to the corresponding baseline. Moreover, within the context of historical heuristics, we identify the Instruction prompt style as the most effective template compared to the InstructionLabel and InstructionMask for LLMs in bug fixing. Finally, we evaluate the cost of HAFix in terms of inference time and token usage, and provide a pragmatic trade-off analysis of the cost and bug-fixing performance, offering valuable insights for the practical deployment of our approach in real-world scenarios.

cs.SE↗

arXiv2026

Permissive-Washing in the Open AI Supply Chain: A Large-Scale Audit of License Integrity

Permissive licenses like MIT, Apache-2.0, and BSD-3-Clause dominate open-source AI, signaling that artifacts like models, datasets, and code can be freely used, modified, and redistributed. However, these licenses carry mandatory requirements: include the full license text, provide a copyright notice, and preserve upstream attribution, that remain unverified at scale. Failure to meet these conditions can place reuse outside the scope of the license, effectively leaving AI artifacts under default copyright for those uses and exposing downstream users to litigation. We call this phenomenon ``permissive washing'': labeling AI artifacts as free to use, while omitting the legal documentation required to make that label actionable. To assess how widespread permissive washing is in the AI supply chain, we empirically audit 124,278 dataset $\rightarrow$ model $\rightarrow$ application supply chains, spanning 3,338 datasets, 6,664 models, and 28,516 applications across Hugging Face and GitHub. We find that an astonishing 96.5\% of datasets and 95.8\% of models lack the required license text, only 2.3\% of datasets and 3.2\% of models satisfy both license text and copyright requirements, and even when upstream artifacts provide complete licensing evidence, attribution rarely propagates downstream: only 27.59\% of models preserve compliant dataset notices and only 5.75\% of applications preserve compliant model notices (with just 6.38\% preserving any linked upstream notice). Practitioners cannot assume permissive labels confer the rights they claim: license files and notices, not metadata, are the source of legal truth. To support future research, we release our full audit dataset and reproducible pipeline.

cs.LG↗

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↗

arXiv2026

Assessing and Improving the Representativeness of Code Generation Benchmarks Using Knowledge Units (KUs) of Programming Languages -- An Empirical Study

Large Language Models (LLMs) such as GPT-4, Claude and LLaMA have shown impressive performance in code generation, typically evaluated using benchmarks (e.g., HumanEval). However, effective code generation requires models to understand and apply a wide range of language concepts. If the concepts exercised in benchmarks are not representative of those used in real-world projects, evaluations may yield incomplete. Despite this concern, the representativeness of code concepts in benchmarks has not been systematically examined. To address this gap, we present the first empirical study that analyzes the representativeness of code generation benchmarks through the lens of Knowledge Units (KUs) - cohesive sets of programming language capabilities provided by language constructs and APIs. We analyze KU coverage in two widely used Python benchmarks, HumanEval and MBPP, and compare them with 30 real-world Python projects. Our results show that each benchmark covers only half of the identified 20 KUs, whereas projects exercise all KUs with relatively balanced distributions. In contrast, benchmark tasks exhibit highly skewed KU distributions. To mitigate this misalignment, we propose a prompt-based LLM framework that synthesizes KU-based tasks to rebalance benchmark KU distributions and better align them with real-world usage. Using this framework, we generate 440 new tasks and augment existing benchmarks. The augmented benchmarks substantially improve KU coverage and achieve over a 60% improvement in distributional alignment. Evaluations of state-of-the-art LLMs on these augmented benchmarks reveal consistent and statistically significant performance drops (12.54-44.82%), indicating that existing benchmarks overestimate LLM performance due to their limited KU coverage. Our findings provide actionable guidance for building more realistic evaluations of LLM code-generation capabilities.

cs.SE↗

arXiv2026

Understanding Prompt Management in GitHub Repositories: A Call for Best Practices

The rapid adoption of foundation models (e.g., large language models) has given rise to promptware, i.e., software built using natural language prompts. Effective management of prompts, such as organization and quality assurance, is essential yet challenging. In this study, we perform an empirical analysis of 24,800 open-source prompts from 92 GitHub repositories to investigate prompt management practices and quality attributes. Our findings reveal critical challenges such as considerable inconsistencies in prompt formatting, substantial internal and external prompt duplication, and frequent readability and spelling issues. Based on these findings, we provide actionable recommendations for developers to enhance the usability and maintainability of open-source prompts within the rapidly evolving promptware ecosystem.

cs.SE↗

arXiv2025

An Empirical Study of Self-Admitted Technical Debt in Machine Learning Software

The emergence of open-source ML libraries such as TensorFlow and Google Auto ML has enabled developers to harness state-of-the-art ML algorithms with minimal overhead. However, during this accelerated ML development process, said developers may often make sub-optimal design and implementation decisions, leading to the introduction of technical debt that, if not addressed promptly, can have a significant impact on the quality of the ML-based software. Developers frequently acknowledge these sub-optimal design and development choices through code comments during software development. These comments, which often highlight areas requiring additional work or refinement in the future, are known as self-admitted technical debt (SATD). This paper aims to investigate SATD in ML code by analyzing 318 open-source ML projects across five domains, along with 318 non-ML projects. We detected SATD in source code comments throughout the different project snapshots, conducted a manual analysis of the identified SATD sample to comprehend the nature of technical debt in the ML code, and performed a survival analysis of the SATD to understand the evolution of such debts. We observed: i) Machine learning projects have a median percentage of SATD that is twice the median percentage of SATD in non-machine learning projects. ii) ML pipeline components for data preprocessing and model generation logic are more susceptible to debt than model validation and deployment components. iii) SATDs appear in ML projects earlier in the development process compared to non-ML projects. iv) Long-lasting SATDs are typically introduced during extensive code changes that span multiple files exhibiting low complexity.

cs.SE↗

arXiv2025

Agentic Refactoring: An Empirical Study of AI Coding Agents

Agentic coding tools, such as OpenAI Codex, Claude Code, and Cursor, are transforming the software engineering landscape. These AI-powered systems function as autonomous teammates capable of planning and executing complex development tasks. Agents have become active participants in refactoring, a cornerstone of sustainable software development aimed at improving internal code quality without altering observable behavior. Despite their increasing adoption, there is a critical lack of empirical understanding regarding how agentic refactoring is utilized in practice, how it compares to human-driven refactoring, and what impact it has on code quality. To address this empirical gap, we present a large-scale study of AI agent-generated refactorings in real-world open-source Java projects, analyzing 15,451 refactoring instances across 12,256 pull requests and 14,988 commits derived from the AIDev dataset. Our empirical analysis shows that refactoring is a common and intentional activity in this development paradigm, with agents explicitly targeting refactoring in 26.1% of commits. Analysis of refactoring types reveals that agentic efforts are dominated by low-level, consistency-oriented edits, such as Change Variable Type (11.8%), Rename Parameter (10.4%), and Rename Variable (8.5%), reflecting a preference for localized improvements over the high-level design changes common in human refactoring. Additionally, the motivations behind agentic refactoring focus overwhelmingly on internal quality concerns, with maintainability (52.5%) and readability (28.1%). Furthermore, quantitative evaluation of code quality metrics shows that agentic refactoring yields small but statistically significant improvements in structural metrics, particularly for medium-level changes, reducing class size and complexity (e.g., Class LOC median $Δ$ = -15.25).

cs.SE↗

arXiv2025

Output Format Biases in the Evaluation of Large Language Models for Code Translation

Code translation between programming languages (PLs) is a critical task in software engineering, facilitating the modernization of legacy systems, ensuring cross-platform compatibility, and enhancing software performance. Most existing studies instruct LLMs to perform code translation and evaluate their performance by either running the generated outputs through test suites or comparing them to reference outputs (ground truth). These outputs, however, may contain not only executable source code but also additional non-code elements, such as natural language explanations or formatting tokens. We refer to the combination of source code and non-code elements as the output format. It is crucial to understand and address variations in output format, as non-code elements can interfere with evaluation metrics, resulting in biased assessments of model performance and comparisons. We conduct an empirical analysis of the outputs from eleven instruct-tuned open-source LLMs, across five PLs: C, C++, Go, Java, and Python. The results show that between 26.4% and 73.7% of outputs produced by our evaluated LLMs necessitate post-processing. To mitigate output format bias, we propose a strategic combination of prompt engineering and regular expressions that effectively extracts source code from mixed-format outputs, enabling the eleven open-source models to achieve an average Code Extraction Success Rate (CSR) of 92.73%. Our empirical study confirms that output format bias affects widely used execution-based metrics, i.e., Computational Accuracy (CA), and text-based metrics, i.e., BLEU, CodeBLEU and CrystalBLEU. Additionally, we test five closed-source LLMs and observe that they also generate varying distributions of output formats, which could lead to output format biases. Our results highlight the need to mitigate the output format bias to enable reliable evaluations in LLMs for code translation.

cs.SE↗

arXiv2025

On the Impact of Black-box Deployment Strategies for Edge AI on Latency and Model Performance

Deciding what combination of operators to use across the Edge AI tiers to achieve specific latency and model performance requirements is an open question for MLOps engineers. This study aims to empirically assess the accuracy vs inference time trade-off of different black-box Edge AI deployment strategies, i.e., combinations of deployment operators and deployment tiers. In this paper, we conduct inference experiments involving 3 deployment operators (i.e., Partitioning, Quantization, Early Exit), 3 deployment tiers (i.e., Mobile, Edge, Cloud) and their combinations on four widely used Computer-Vision models to investigate the optimal strategies from the point of view of MLOps developers. Our findings suggest that Edge deployment using the hybrid Quantization + Early Exit operator could be preferred over non-hybrid operators (Quantization/Early Exit on Edge, Partition on Mobile-Edge) when faster latency is a concern at medium accuracy loss. However, when minimizing accuracy loss is a concern, MLOps engineers should prefer using only a Quantization operator on edge at a latency reduction or increase, respectively over the Early Exit/Partition (on edge/mobile-edge) and Quantized Early Exit (on edge) operators. In scenarios constrained by Mobile CPU/RAM resources, a preference for Partitioning across mobile and edge tiers is observed over mobile deployment. For models with smaller input data samples (such as FCN), a network-constrained cloud deployment can also be a better alternative than Mobile/Edge deployment and Partitioning strategies. For models with large input data samples (ResNet, ResNext, DUC), an edge tier having higher network/computational capabilities than Cloud/Mobile can be a more viable option than Partitioning and Mobile/Cloud deployment strategies.

cs.SE↗

arXiv2025

OmniLLP: Enhancing LLM-based Log Level Prediction with Context-Aware Retrieval

Developers insert logging statements in source code to capture relevant runtime information essential for maintenance and debugging activities. Log level choice is an integral, yet tricky part of the logging activity as it controls log verbosity and therefore influences systems' observability and performance. Recent advances in ML-based log level prediction have leveraged large language models (LLMs) to propose log level predictors (LLPs) that demonstrated promising performance improvements (AUC between 0.64 and 0.8). Nevertheless, current LLM-based LLPs rely on randomly selected in-context examples, overlooking the structure and the diverse logging practices within modern software projects. In this paper, we propose OmniLLP, a novel LLP enhancement framework that clusters source files based on (1) semantic similarity reflecting the code's functional purpose, and (2) developer ownership cohesion. By retrieving in-context learning examples exclusively from these semantic and ownership aware clusters, we aim to provide more coherent prompts to LLPs leveraging LLMs, thereby improving their predictive accuracy. Our results show that both semantic and ownership-aware clusterings statistically significantly improve the accuracy (by up to 8\% AUC) of the evaluated LLM-based LLPs compared to random predictors (i.e., leveraging randomly selected in-context examples from the whole project). Additionally, our approach that combines the semantic and ownership signal for in-context prediction achieves an impressive 0.88 to 0.96 AUC across our evaluated projects. Our findings highlight the value of integrating software engineering-specific context, such as code semantic and developer ownership signals into LLM-LLPs, offering developers a more accurate, contextually-aware approach to logging and therefore, enhancing system maintainability and observability.

cs.SE↗

arXiv2025

Blended PC Peer Review Model: Process and Reflection

The academic peer review system is under increasing pressure due to a growing volume of submissions and a limited pool of available reviewers, resulting in delayed decisions and an uneven distribution of reviewing responsibilities. Building upon the International Conference on Mining Software Repositories (MSR) community's earlier experience with a Shadow PC (2021 and 2022) and Junior PC (2023 and 2024), MSR 2025 experimented with a Blended Program Committee (PC) peer review model for its Technical Track. This new model pairs up one Junior PC member with two regular PC members as part of the core review team of a given paper, instead of adding them as an extra reviewer. This paper presents the rationale, implementation, and reflections on the model, including empirical insights from a post-review author survey evaluating the quality and usefulness of reviews. Our findings highlight the potential of a Blended PC to alleviate reviewer shortages, foster inclusivity, and sustain a high-quality peer review process. We offer lessons learned and recommendations to guide future adoption and refinement of the model.

cs.SE↗

arXiv2025

SwarmUpdate: Hierarchical Software Updates and Deep Learning Model Patching for Heterogeneous UAV Swarms

Heterogeneous unmanned aerial vehicle (UAV) swarms consist of dozens to hundreds of drones with different roles and varying hardware and software requirements collaborating towards a shared mission. While traditional approaches for synchronized software updates assume swarms to be unstructured and homogeneous, the heterogeneous nature of modern swarms and the emerging need of drones to update their deep learning (perception) models with new objectives or data as a mission unfolds, has made efficient software update methods crucial for swarms to adapt to dynamic environments. To address these challenges, we introduce the SwarmUpdate framework for software updates in heterogeneous UAV swarms, composed of two key components: SwarmSync and SwarmModelPatch. SwarmSync is a hierarchical software update synchronization strategy to distribute a software update to the right subset of drones within a swarm, while SwarmModelPatch is a deep learning model patching method that reduces the size of a (deep learning model) update by only allowing some layers of the model to be updated (freezing the other layers). In this paper, we systematically evaluate the performance of SwarmSync through large-scale simulations in the ARGoS swarm simulator, comparing SwarmSync to auction-based (SOUL) and gossip-based rebroadcasting (Gossip) baselines, and SwarmModelPatch to a non-incremental model patching strategy.

cs.SE↗

arXiv2025

Towards Semantic Versioning of Open Pre-trained Language Model Releases on Hugging Face

The proliferation of open Pre-trained Language Models (PTLMs) on model registry platforms like Hugging Face (HF) presents both opportunities and challenges for companies building products around them. Similar to traditional software dependencies, PTLMs continue to evolve after a release. However, the current state of release practices of PTLMs on model registry platforms are plagued by a variety of inconsistencies, such as ambiguous naming conventions and inaccessible model training documentation. Given the knowledge gap on current PTLM release practices, our empirical study uses a mixed-methods approach to analyze the releases of 52,227 PTLMs on the most well-known model registry, HF. Our results reveal 148 different naming practices for PTLM releases, with 40.87% of changes to model weight files not represented in the adopted name-based versioning practice or their documentation. In addition, we identified that the 52,227 PTLMs are derived from only 299 different base models (the modified original models used to create 52,227 PTLMs), with Fine-tuning and Quantization being the most prevalent modification methods applied to these base models. Significant gaps in release transparency, in terms of training dataset specifications and model card availability, still exist, highlighting the need for standardized documentation. While we identified a model naming practice explicitly differentiating between major and minor PTLM releases, we did not find any significant difference in the types of changes that went into either type of releases, suggesting that major/minor version numbers for PTLMs often are chosen arbitrarily. Our findings provide valuable insights to improve PTLM release practices, nudging the field towards more formal semantic versioning practices.

cs.SE↗