Daily Papers

Interest in applying Artificial Intelligence (AI) techniques to compiler optimizations is increasing rapidly, but compiler research has a high entry barrier. Unlike in other domains, compiler and AI researchers do not have access to the datasets and frameworks that enable fast iteration and development of ideas, and getting started requires a significant engineering investment. What is needed is an easy, reusable experimental infrastructure for real world compiler optimization tasks that can serve as a common benchmark for comparing techniques, and as a platform to accelerate progress in the field. We introduce CompilerGym, a set of environments for real world compiler optimization tasks, and a toolkit for exposing new optimization tasks to compiler researchers. CompilerGym enables anyone to experiment on production compiler optimization problems through an easy-to-use package, regardless of their experience with compilers. We build upon the popular OpenAI Gym interface enabling researchers to interact with compilers using Python and a familiar API. We describe the CompilerGym architecture and implementation, characterize the optimization spaces and computational efficiencies of three included compiler environments, and provide extensive empirical evaluations. Compared to prior works, CompilerGym offers larger datasets and optimization spaces, is 27x more computationally efficient, is fault-tolerant, and capable of detecting reproducibility bugs in the underlying compilers. In making it easy for anyone to experiment with compilers - irrespective of their background - we aim to accelerate progress in the AI and compiler research domains.

Software developers attempt to reproduce software bugs to understand their erroneous behaviours and to fix them. Unfortunately, they often fail to reproduce (or fix) them, which leads to faulty, unreliable software systems. However, to date, only a little research has been done to better understand what makes the software bugs non-reproducible. In this paper, we conduct a multimodal study to better understand the non-reproducibility of software bugs. First, we perform an empirical study using 576 non-reproducible bug reports from two popular software systems (Firefox, Eclipse) and identify 11 key factors that might lead a reported bug to non-reproducibility. Second, we conduct a user study involving 13 professional developers where we investigate how the developers cope with non-reproducible bugs. We found that they either close these bugs or solicit for further information, which involves long deliberations and counter-productive manual searches. Third, we offer several actionable insights on how to avoid non-reproducibility (e.g., false-positive bug report detector) and improve reproducibility of the reported bugs (e.g., sandbox for bug reproduction) by combining our analyses from multiple studies (e.g., empirical study, developer study).

Bug-fix benchmarks are essential for evaluating methodologies in automatic program repair (APR) and fault localization (FL). However, existing benchmarks, exemplified by Defects4J, need to evolve to incorporate recent bug-fixes aligned with contemporary development practices. Moreover, reproducibility, a key scientific principle, has been lacking in bug-fix benchmarks. To address these gaps, we present GitBug-Java, a reproducible benchmark of recent Java bugs. GitBug-Java features 199 bugs extracted from the 2023 commit history of 55 notable open-source repositories. The methodology for building GitBug-Java ensures the preservation of bug-fixes in fully-reproducible environments. We publish GitBug-Java at https://github.com/gitbugactions/gitbug-java.

Despite its crucial role in research experiments, code correctness is often presumed only on the basis of the perceived quality of results. This assumption comes with the risk of erroneous outcomes and potentially misleading findings. To address this issue, we posit that the current focus on reproducibility should go hand in hand with the emphasis on software quality. We present a case study in which we identify and fix three bugs in widely used implementations of the state-of-the-art Conformer architecture. Through experiments on speech recognition and translation in various languages, we demonstrate that the presence of bugs does not prevent the achievement of good and reproducible results, which however can lead to incorrect conclusions that potentially misguide future research. As a countermeasure, we propose a Code-quality Checklist and release pangoliNN, a library dedicated to testing neural models, with the goal of promoting coding best practices and improving research software quality within the NLP community.

Jupyter notebooks facilitate the bundling of executable code with its documentation and output in one interactive environment, and they represent a popular mechanism to document and share computational workflows. The reproducibility of computational aspects of research is a key component of scientific reproducibility but has not yet been assessed at scale for Jupyter notebooks associated with biomedical publications. We address computational reproducibility at two levels: First, using fully automated workflows, we analyzed the computational reproducibility of Jupyter notebooks related to publications indexed in PubMed Central. We identified such notebooks by mining the articles full text, locating them on GitHub and re-running them in an environment as close to the original as possible. We documented reproduction success and exceptions and explored relationships between notebook reproducibility and variables related to the notebooks or publications. Second, this study represents a reproducibility attempt in and of itself, using essentially the same methodology twice on PubMed Central over two years. Out of 27271 notebooks from 2660 GitHub repositories associated with 3467 articles, 22578 notebooks were written in Python, including 15817 that had their dependencies declared in standard requirement files and that we attempted to re-run automatically. For 10388 of these, all declared dependencies could be installed successfully, and we re-ran them to assess reproducibility. Of these, 1203 notebooks ran through without any errors, including 879 that produced results identical to those reported in the original notebook and 324 for which our results differed from the originally reported ones. Running the other notebooks resulted in exceptions. We zoom in on common problems, highlight trends and discuss potential improvements to Jupyter-related workflows associated with biomedical publications.

One of the challenges in machine learning research is to ensure that presented and published results are sound and reliable. Reproducibility, that is obtaining similar results as presented in a paper or talk, using the same code and data (when available), is a necessary step to verify the reliability of research findings. Reproducibility is also an important step to promote open and accessible research, thereby allowing the scientific community to quickly integrate new findings and convert ideas to practice. Reproducibility also promotes the use of robust experimental workflows, which potentially reduce unintentional errors. In 2019, the Neural Information Processing Systems (NeurIPS) conference, the premier international conference for research in machine learning, introduced a reproducibility program, designed to improve the standards across the community for how we conduct, communicate, and evaluate machine learning research. The program contained three components: a code submission policy, a community-wide reproducibility challenge, and the inclusion of the Machine Learning Reproducibility checklist as part of the paper submission process. In this paper, we describe each of these components, how it was deployed, as well as what we were able to learn from this initiative.

Bug reports often lack sufficient detail for developers to reproduce and fix the underlying defects. Bug Reproduction Tests (BRTs), tests that fail when the bug is present and pass when it has been resolved, are crucial for debugging, but they are rarely included in bug reports, both in open-source and in industrial settings. Thus, automatically generating BRTs from bug reports has the potential to accelerate the debugging process and lower time to repair. This paper investigates automated BRT generation within an industry setting, specifically at Google, focusing on the challenges of a large-scale, proprietary codebase and considering real-world industry bugs extracted from Google's internal issue tracker. We adapt and evaluate a state-of-the-art BRT generation technique, LIBRO, and present our agent-based approach, BRT Agent, which makes use of a fine-tuned Large Language Model (LLM) for code editing. Our BRT Agent significantly outperforms LIBRO, achieving a 28% plausible BRT generation rate, compared to 10% by LIBRO, on 80 human-reported bugs from Google's internal issue tracker. We further investigate the practical value of generated BRTs by integrating them with an Automated Program Repair (APR) system at Google. Our results show that providing BRTs to the APR system results in 30% more bugs with plausible fixes. Additionally, we introduce Ensemble Pass Rate (EPR), a metric which leverages the generated BRTs to select the most promising fixes from all fixes generated by APR system. Our evaluation on EPR for Top-K and threshold-based fix selections demonstrates promising results and trade-offs. For example, EPR correctly selects a plausible fix from a pool of 20 candidates in 70% of cases, based on its top-1 ranking.

Concerns about reproducibility in artificial intelligence (AI) have emerged, as researchers have reported unsuccessful attempts to directly reproduce published findings in the field. Replicability, the ability to affirm a finding using the same procedures on new data, has not been well studied. In this paper, we examine both reproducibility and replicability of a corpus of 16 papers on table structure recognition (TSR), an AI task aimed at identifying cell locations of tables in digital documents. We attempt to reproduce published results using codes and datasets provided by the original authors. We then examine replicability using a dataset similar to the original as well as a new dataset, GenTSR, consisting of 386 annotated tables extracted from scientific papers. Out of 16 papers studied, we reproduce results consistent with the original in only four. Two of the four papers are identified as replicable using the similar dataset under certain IoU values. No paper is identified as replicable using the new dataset. We offer observations on the causes of irreproducibility and irreplicability. All code and data are available on Codeocean at https://codeocean.com/capsule/6680116/tree.

Reproducibility is central to research credibility, yet large-scale reanalysis of empricial data remains costly because replication packages vary widely in structure, software environment, and documentation. We develop and evaluate an agentic AI workflow that addresses this execution bottleneck while preserving scientific rigor. The system separates scientific reasoning from computational execution: researchers design fixed diagnostic templates, and the workflow automates the acquisition, harmonization, and execution of replication materials using pre-specified, version-controlled code. A structured knowledge layer records resolved failure patterns, enabling adaptation across heterogeneous studies while keeping each pipeline version transparent and stable. We evaluate this workflow on 92 instrumental variable (IV) studies, including 67 with manually verified reproducible 2SLS estimates and 25 newly published IV studies under identical criteria. For each paper, we analyze up to three two-stage least squares (2SLS) specifications, totaling 215. Across the 92 papers, the system achieves 87% end-to-end success overall. Conditional on accessible data and code, reproducibility is 100% at both the paper and specification levels. The framework substantially lowers the cost of executing established empirical protocols and can be adapted in empirical settings where analytic templates and norms of transparency are well established.

In recent years, significant progress has been made in solving challenging problems across various domains using deep reinforcement learning (RL). Reproducing existing work and accurately judging the improvements offered by novel methods is vital to sustaining this progress. Unfortunately, reproducing results for state-of-the-art deep RL methods is seldom straightforward. In particular, non-determinism in standard benchmark environments, combined with variance intrinsic to the methods, can make reported results tough to interpret. Without significance metrics and tighter standardization of experimental reporting, it is difficult to determine whether improvements over the prior state-of-the-art are meaningful. In this paper, we investigate challenges posed by reproducibility, proper experimental techniques, and reporting procedures. We illustrate the variability in reported metrics and results when comparing against common baselines and suggest guidelines to make future results in deep RL more reproducible. We aim to spur discussion about how to ensure continued progress in the field by minimizing wasted effort stemming from results that are non-reproducible and easily misinterpreted.

Reproducing research results from papers and released code is central to scientific progress. Existing works have introduced benchmarks to evaluate whether LLM agents can assist with reproducibility, but they are difficult to scale due to their reliance on substantial manual effort for data curation and evaluation. We introduce ReproRepo, a scalable framework for reproducibility evaluation that leverages human-raised GitHub issues as naturally occurring supervision on realistic reproduction blockers. We instantiate ReproRepo on 1,149 recent machine learning papers from major conferences and evaluate four frontier model-agent configurations. Our results show that LLM agents, even without executing code, can identify many real-world reproducibility problems from paper-repository pairs: the best agent in our study, namely Codex with GPT-5.5, surfaces at least one semantically related human-reported blocker for ~90% of papers in the study. Further analysis shows that agents are particularly effective for surfacing visible failures and identifying the right semantic region, but may still be insufficient in exact localization. ReproRepo can serve as a reusable, scalable framework for future evaluations of LLM agents on real-world reproducibility auditing. Our code is released at https://github.com/LithiumDA/ReproRepo.

Reproducing buggy code is the first and crucially important step in issue resolving, as it aids in identifying the underlying problems and validating that generated patches resolve the problem. While numerous approaches have been proposed for this task, they primarily address common, widespread errors and struggle to adapt to unique, evolving errors specific to individual code repositories. To fill this gap, we propose EvoCoder, a multi-agent continuous learning framework for issue code reproduction. EvoCoder adopts a reflection mechanism that allows the LLM to continuously learn from previously resolved problems and dynamically refine its strategies to new emerging challenges. To prevent experience bloating, EvoCoder introduces a novel hierarchical experience pool that enables the model to adaptively update common and repo-specific experiences. Our experimental results show a 20\% improvement in issue reproduction rates over existing SOTA methods. Furthermore, integrating our reproduction mechanism significantly boosts the overall accuracy of the existing issue-resolving pipeline.

Reproducing game bugs, particularly crash bugs in continuously evolving games like Minecraft, is a notoriously manual, time-consuming, and challenging process to automate; insights from a key decision maker from Minecraft we interviewed confirm this, highlighting that a substantial portion of crash reports necessitate manual scenario reconstruction. Despite the success of LLM-driven bug reproduction in other software domains, games, with their complex interactive environments, remain largely unaddressed. This paper introduces BugCraft, a novel end-to-end framework designed to automate the reproduction of crash bugs in Minecraft directly from user-submitted bug reports, addressing the critical gap in automated game bug reproduction. BugCraft employs a two-stage approach: first, a Step Synthesizer leverages LLMs and Minecraft Wiki knowledge to transform bug reports into high-quality, structured steps to reproduce (S2R). Second, an Action Model, powered by a vision-based LLM agent and a custom macro API, executes these S2R steps within Minecraft to trigger the reported crash. To facilitate evaluation, we introduce BugCraft-Bench, a curated dataset of Minecraft crash bug reports. On BugCraft-Bench, our framework end-to-end reproduced 34.9% of crash bugs with GPT-4.1, outperforming baseline computer-use models by 37%. BugCraft demonstrates the feasibility of automated reproduction of crash bugs in complex game environments using LLMs, opening promising avenues for game testing and development. Finally, we make our code open at https://bugcraft2025.github.io

In software development, bug report reproduction is a challenging task. This paper introduces ReBL, a novel feedback-driven approach that leverages GPT-4, a large-scale language model, to automatically reproduce Android bug reports. Unlike traditional methods, ReBL bypasses the use of Step to Reproduce (S2R) entities. Instead, it leverages the entire textual bug report and employs innovative prompts to enhance GPT's contextual reasoning. This approach is more flexible and context-aware than the traditional step-by-step entity matching approach, resulting in improved accuracy and effectiveness. In addition to handling crash reports, ReBL has the capability of handling non-crash bug reports. Our evaluation of 96 Android bug reports (73 crash and 23 non-crash) demonstrates that ReBL successfully reproduced 90.63% of these reports, averaging only 74.98 seconds per bug report. Additionally, ReBL outperformed three existing tools in both success rate and speed.

There are various benchmarks to evaluate bugfixing capabilities of Large Language Models. However, most widespread benchmarks do not fully reflect real-world bugfixing practices. They are small, weakening statistical reliability, and the buggy programs are often similar to one another, potentially distorting evaluation results. The range of bug types can also be narrow, failing to capture a representative range of bugs. To address these issues, we introduce MegaBugFix, a large-scale bugfixing benchmark containing 12,629 buggy Python programs synthesized from correct ones by a Large Language Model. Bug injections were generated as diffs representing code changes. Through this approach, we were able to avoid common pitfalls of LLM-based mutation techniques like injecting overly simplistic bugs or failing to modify the input program. We evaluated 13 open-weight models on MegaBugFix and baseline benchmarks, finding consistently lower performance on MegaBugFix. This reveals that our benchmark presents more challenging bugs and exposes model failures that may remain hidden when evaluating on existing benchmarks. The benchmark and fine-tuned model used for bug injection are available at hf.co/collections/szalontaib/megabugfix.

The literature has witnessed an emerging interest in AI agents for automated assessment of scientific papers. Existing benchmarks focus primarily on the computational aspect of this task, testing agents' ability to reproduce or replicate research outcomes when having access to the code and data. This setting, while foundational, (1) fails to capture the inconsistent availability of new data for replication as opposed to reproduction, and (2) lacks ground-truth diversity by focusing only on reproducible papers, thereby failing to evaluate an agent's ability to identify non-replicable research. Furthermore, most benchmarks only evaluate outcomes rather than the replication process. In response, we introduce ReplicatorBench, an end-to-end benchmark, including human-verified replicable and non-replicable research claims in social and behavioral sciences for evaluating AI agents in research replication across three stages: (1) extraction and retrieval of replication data; (2) design and execution of computational experiments; and (3) interpretation of results, allowing a test of AI agents' capability to mimic the activities of human replicators in real world. To set a baseline of AI agents' capability, we develop ReplicatorAgent, an agentic framework equipped with necessary tools, like web search and iterative interaction with sandboxed environments, to accomplish tasks in ReplicatorBench. We evaluate ReplicatorAgent across four underlying large language models (LLMs), as well as different design choices of programming language and levels of code access. Our findings reveal that while current LLM agents are capable of effectively designing and executing computational experiments, they struggle with retrieving resources, such as new data, necessary to replicate a claim. All code and data are publicly available at https://github.com/CenterForOpenScience/llm-benchmarking.

Efficient experiment reproduction is critical to accelerating progress in artificial intelligence. However, the inherent complexity of method design and training procedures presents substantial challenges for automation. Notably, reproducing experiments often requires implicit domain-specific knowledge not explicitly documented in the original papers. To address this, we introduce the paper lineage algorithm, which identifies and extracts implicit knowledge from the relevant references cited by the target paper. Building on this idea, we propose AutoReproduce, a multi-agent framework capable of automatically reproducing experiments described in research papers in an end-to-end manner. AutoReproduce enhances code executability by generating unit tests alongside the reproduction process. To evaluate the reproduction capability, we construct ReproduceBench, a benchmark annotated with verified implementations, and introduce novel evaluation metrics to assess both the reproduction and execution fidelity. Experimental results demonstrate that AutoReproduce outperforms the existing strong agent baselines on all five evaluation metrics by a peak margin of over 70%. In particular, compared to the official implementations, AutoReproduce achieves an average performance gap of 22.1% on 89.74% of the executable experiment runs. The code will be available at https://github.com/AI9Stars/AutoReproduce.

Bug-fix benchmarks are fundamental in advancing various sub-fields of software engineering such as automatic program repair (APR) and fault localization (FL). A good benchmark must include recent examples that accurately reflect technologies and development practices of today. To be executable in the long term, a benchmark must feature test suites that do not degrade overtime due to, for example, dependencies that are no longer available. Existing benchmarks fail in meeting both criteria. For instance, Defects4J, one of the foremost Java benchmarks, last received an update in 2020. Moreover, full-reproducibility has been neglected by the majority of existing benchmarks. In this paper, we present GitBug-Actions: a novel tool for building bug-fix benchmarks with modern and fully-reproducible bug-fixes. GitBug-Actions relies on the most popular CI platform, GitHub Actions, to detect bug-fixes and smartly locally execute the CI pipeline in a controlled and reproducible environment. To the best of our knowledge, we are the first to rely on GitHub Actions to collect bug-fixes. To demonstrate our toolchain, we deploy GitBug-Actions to build a proof-of-concept Go bug-fix benchmark containing executable, fully-reproducible bug-fixes from different repositories. A video demonstrating GitBug-Actions is available at: https://youtu.be/aBWwa1sJYBs.

Reproducibility is essential for scientific research. However, in computer vision, achieving consistent results is challenging due to various factors. One influential, yet often unrecognized, factor is CUDA-induced randomness. Despite CUDA's advantages for accelerating algorithm execution on GPUs, if not controlled, its behavior across multiple executions remains non-deterministic. While reproducibility issues in ML being researched, the implications of CUDA-induced randomness in application are yet to be understood. Our investigation focuses on this randomness across one standard benchmark dataset and two real-world datasets in an isolated environment. Our results show that CUDA-induced randomness can account for differences up to 4.77% in performance scores. We find that managing this variability for reproducibility may entail increased runtime or reduce performance, but that disadvantages are not as significant as reported in previous studies.

High quality bugs are key to training the next generation of language model based software engineering (SWE) agents. We introduce a novel method for synthetic generation of difficult and diverse bugs. Our method instructs SWE Agents to introduce a feature into the codebase whereby they may unintentionally break tests, resulting in bugs. Prior approaches often induce an out-of-distribution effect by generating bugs intentionally (e.g. by introducing local perturbation to existing code), which does not reflect realistic development processes. We perform qualitative analysis to demonstrate that our approach for generating bugs more closely reflects the patterns found in human-authored edits. Through extensive experiments, we demonstrate that our bugs provide more efficient training data for supervised fine-tuning, outperforming other bug datasets by 2% with half the training data (1.2k vs. 3k bugs). We train on our newly generated bugs in addition to existing bug datasets to get FrogBoss a state-of-the-art 32B parameter model on SWE-bench Verified with a pass@1 of 54.6% and FrogMini a state-of-the-art 14B model on SWE-bench Verified with a pass@1 of 45.3% on SWE-bench Verified averaged over three seeds.

Unlike code completion, debugging requires localizing faults and applying targeted edits. We observe that frontier LLMs often regenerate correct but over-edited solutions during debugging. To evaluate how far LLMs are from precise debugging, we introduce the Precise Debugging Benchmark (PDB) framework, which automatically converts any coding dataset into a debugging benchmark with precision-aware evaluation. PDB generates buggy programs by synthesizing verified atomic bugs and composing them into multi-bug programs. We define two novel metrics, edit-level precision and bug-level recall, which measures how many necessary edits are made and how many bugs are resolved. We release two evaluation benchmarks: PDB-Single-Hard on single-line bugs, and PDB-Multi on multi-line bugs. Experiments show that frontier models, such as GPT-5.1-Codex and DeepSeek-V3.2-Thinking, achieve unit-test pass rates above 76% but exhibit precision below 45%, even when explicitly instructed to perform minimal debugging. Finally, we show that iterative and agentic debugging strategies do not substantially improve precision or recall, highlighting the need to rethink post-training pipelines for coding models.

Large language model (LLM)-based software engineering agents are increasingly developed to resolve software issues by generating patches from issue reports and code repositories. Bug reproduction tests (BRTs) are an important building block for such agents and have been shown useful for patch validation. However, it remains unclear whether BRTs can also help the more central stage of patch generation. We first conduct a preliminary study and find that directly using advanced BRT generators to guide patch generation is not beneficial: fail-to-fail BRTs can mislead agents, while even fail-to-pass BRTs bring limited or negative gains. Our analysis reveals two reasons: fail-to-pass BRTs may cover only one manifestation of the reported issue, leading to partial patches, whereas fail-to-fail BRTs are unreliable as direct patch-generation targets. Motivated by these insights, we propose SWE-Doctor, a software issue resolution agent that guides patch generation with runtime diagnoses derived from multi-faceted BRT executions. SWE-Doctor first generates multi-faceted BRTs for different behavioral requirements stated in the issue, then executes and debugs these BRTs to construct runtime-grounded diagnosis records, and finally uses the diagnoses together with localization information inferred during BRT generation to guide patch generation and reduce partial patches. We evaluate SWE-Doctor on Python bug-fixing issues from the widely adopted SWE-bench Verified and SWE-bench Pro across five LLM backends. SWE-Doctor consistently outperforms existing agents across all 10 LLM-benchmark combinations, achieving average resolution rates of 75.7% on SWE-bench Verified and 59.4% on SWE-bench Pro. In particular, on the more challenging SWE-bench Pro, SWE-Doctor improves the average resolution rate by 8.0-8.9 percentage points over the baseline agents.

Assessing the reproducibility of social science papers is essential for promoting rigor in research processes, but manual assessment is costly. With recent advances in agentic AI systems (i.e., AI agents), we seek to evaluate their capability to automate this process. However, existing benchmarks for reproducing research papers (1) focus solely on reproducing results using provided code and data without assessing their consistency with the paper, (2) oversimplify real-world scenarios, and (3) lack necessary diversity in data formats and programming languages. To address these issues, we introduce REPRO-Bench, a collection of 112 task instances, each representing a social science paper with a publicly available reproduction report. The agents are tasked with assessing the reproducibility of the paper based on the original paper PDF and the corresponding reproduction package. REPRO-Bench features end-to-end evaluation tasks on the reproducibility of social science papers with complexity comparable to real-world assessments. We evaluate three representative AI agents on REPRO-Bench, with the best-performing agent achieving an accuracy of only 21.4%. Building on our empirical analysis, we develop REPRO-Agent, which improves the highest accuracy achieved by existing agents by 71%. We conclude that more advanced AI agents should be developed to automate real-world reproducibility assessment. REPRO-Bench is publicly available at https://github.com/uiuc-kang-lab/REPRO-Bench.

Reproducibility crises across sciences highlight the limitations of the paper-centric review system in assessing the rigor and reproducibility of research. AI agents that autonomously design and generate large volumes of research outputs exacerbate these challenges. In this work, we address the growing challenges of scalability and rigor by flipping the dynamic and developing AI agents as research evaluators. We propose the first execution-grounded evaluation framework that verifies research beyond narrative review by examining code and data alongside the paper. We use mechanistic interpretability research as a testbed, build standardized research output, and develop MechEvalAgent, an automated evaluation framework that assesses the coherence of the experimental process, the reproducibility of results, and the generalizability of findings. We show that our framework achieves above 80% agreement with human judges, identifies substantial methodological problems, and surfaces 51 additional issues that human reviewers miss. Our work demonstrates the potential of AI agents to transform research evaluation and pave the way for rigorous scientific practices.

AI scientist systems, capable of autonomously executing the full research workflow from hypothesis generation and experimentation to paper writing, hold significant potential for accelerating scientific discovery. However, the internal workflow of these systems have not been closely examined. This lack of scrutiny poses a risk of introducing flaws that could undermine the integrity, reliability, and trustworthiness of their research outputs. In this paper, we identify four potential failure modes in contemporary AI scientist systems: inappropriate benchmark selection, data leakage, metric misuse, and post-hoc selection bias. To examine these risks, we design controlled experiments that isolate each failure mode while addressing challenges unique to evaluating AI scientist systems. Our assessment of two prominent open-source AI scientist systems reveals the presence of several failures, across a spectrum of severity, which can be easily overlooked in practice. Finally, we demonstrate that access to trace logs and code from the full automated workflow enables far more effective detection of such failures than examining the final paper alone. We thus recommend journals and conferences evaluating AI-generated research to mandate submission of these artifacts alongside the paper to ensure transparency, accountability, and reproducibility.

In software maintenance, bug reproduction is essential for effective fault localization and repair. Manually writing reproduction scripts is a time-consuming task with high requirements for developers. Hence, automation of bug reproduction has increasingly attracted attention from researchers and practitioners. However, the existing studies on bug reproduction are generally limited to specific bug types such as program crashes, and hard to be applied to general bug reproduction. In this paper, considering the superior performance of agent-based methods in code intelligence tasks, we focus on designing an agent-based framework for the task. Directly employing agents would lead to limited bug reproduction performance, due to entangled subtasks, lengthy retrieved context, and unregulated actions. To mitigate the challenges, we propose an Automated gEneral buG reproductIon Scripts generation framework, named AEGIS, which is the first agent-based framework for the task. AEGIS mainly contains two modules: (1) A concise context construction module, which aims to guide the code agent in extracting structured information from issue descriptions, identifying issue-related code with detailed explanations, and integrating these elements to construct the concise context; (2) A FSM-based multi-feedback optimization module to further regulate the behavior of the code agent within the finite state machine (FSM), ensuring a controlled and efficient script generation process based on multi-dimensional feedback. Extensive experiments on the public benchmark dataset show that AEGIS outperforms the state-of-the-art baseline by 23.0% in F->P metric. In addition, the bug reproduction scripts generated by AEGIS can improve the relative resolved rate of Agentless by 12.5%.

Automated program repair (APR) agents are transitioning from research benchmarks to developer workflows, yet they still begin with bug reports written for human developers. While decades of research have established what makes a good bug report for humans (e.g., steps to reproduce, stack traces), it remains unclear whether these features transfer to LLM-based agents. We study this question in two analyses. First, we use statistical modeling to examine associations between 27 bug-report features and repair success across 433 SWE-bench Verified issues attempted by 87 repair agents. We find that fix suggestions, reproduction scripts, repository source code, and localization info are associated with higher resolution likelihood, while longer reports are associated with lower odds. Second, we conduct controlled ablations across 2 models and 17 problem-statement mutations on SWE-bench Pro, varying the information available to an agent while holding the underlying task fixed. We remove or isolate selected bug-report content, delete fault-localization cues, and test structural changes that flatten lists or remove section headers. We find that both models depend on localization cues and expected behavior, and that structural changes alone can reduce solve rates, even without removing any content. The two models diverge in how they handle missing information: Qwen searches more widely and can exhaust its turn budget, while Gemma commits to a plausible interpretation early and patches on it. Our findings indicate that a good bug report for an agent overlaps with, but is not identical to, a good report for a human: agents benefit most from concrete, executable, and well-localized information, whereas some qualities long emphasized for human readers, such as natural language steps to reproduce and readable descriptions, contribute little or even correlate with lower success.

Research is facing a reproducibility crisis, in which the results and findings of many studies are difficult or even impossible to reproduce. This is also the case in machine learning (ML) and artificial intelligence (AI) research. Often, this is the case due to unpublished data and/or source-code, and due to sensitivity to ML training conditions. Although different solutions to address this issue are discussed in the research community such as using ML platforms, the level of reproducibility in ML-driven research is not increasing substantially. Therefore, in this mini survey, we review the literature on reproducibility in ML-driven research with three main aims: (i) reflect on the current situation of ML reproducibility in various research fields, (ii) identify reproducibility issues and barriers that exist in these research fields applying ML, and (iii) identify potential drivers such as tools, practices, and interventions that support ML reproducibility. With this, we hope to contribute to decisions on the viability of different solutions for supporting ML reproducibility.

LLM agents span command-line interfaces (e.g., Codex) and agent frameworks (e.g., LangChain), integrating backend LLMs with harness code that parses model outputs, controls agent loops, and manages context. Both the harness and LLM-generated responses jointly shape an agent's execution. This architecture gives rise to bugs that cannot be readily understood by inspecting either component alone, because some bugs occur only when a particular LLM response elicits an abnormal reaction from the agent. Prior empirical studies of agent bugs have largely attributed failures either to limited model capabilities or to harness-side defects, such as outdated APIs and configuration misalignment, without characterizing these AR bugs. We conduct the first empirical study focused on agent-reactive (AR) bugs. Through manual analysis of 255 bug reports from Codex, Gemini-CLI, LangChain, and CrewAI, we construct a two-axis taxonomy covering observable symptoms and the LLM behaviors that trigger them. Our findings show that many AR bugs manifest as silent errors without well-defined test oracles, which makes detection difficult. The stochasticity of LLM responses further complicates bug reproduction. We additionally examine fixes proposed by users and implemented by developers. This analysis exposes a mismatch: users frequently advocate harness-side guardrails, whereas developers may attribute the issue to the LLM or respond slowly to user-proposed fixes. These findings point to the need for mechanisms that help users and developers understand the root causes and resolutions of AR bugs. Overall, the study highlights challenges specific to LLM agents and motivates the design of test oracles, reproduction support, and fault-localization techniques for AR bugs.

Distributed Deep Reinforcement Learning (DRL) aims to leverage more computational resources to train autonomous agents with less training time. Despite recent progress in the field, reproducibility issues have not been sufficiently explored. This paper first shows that the typical actor-learner framework can have reproducibility issues even if hyperparameters are controlled. We then introduce Cleanba, a new open-source platform for distributed DRL that proposes a highly reproducible architecture. Cleanba implements highly optimized distributed variants of PPO and IMPALA. Our Atari experiments show that these variants can obtain equivalent or higher scores than strong IMPALA baselines in moolib and torchbeast and PPO baseline in CleanRL. However, Cleanba variants present 1) shorter training time and 2) more reproducible learning curves in different hardware settings. Cleanba's source code is available at https://github.com/vwxyzjn/cleanba

Efficient bug resolution is critical for maintaining software quality and user satisfaction. However, specific bug reports experience unusually long resolution times, which may indicate underlying process inefficiencies or complex issues. This study presents a comprehensive analysis of bug resolution anomalies across seven prominent open-source repositories: Cassandra, Firefox, Hadoop, HBase, SeaMonkey, Spark, and Thunderbird. Utilizing statistical methods such as Z-score and Interquartile Range (IQR), we identify anomalies in bug resolution durations. To understand the thematic nature of these anomalies, we apply Term Frequency-Inverse Document Frequency (TF-IDF) for textual feature extraction and KMeans clustering to group similar bug summaries. Our findings reveal consistent patterns across projects, with anomalies often clustering around test failures, enhancement requests, and user interface issues. This approach provides actionable insights for project maintainers to prioritize and effectively address long-standing bugs.

As computational work becomes more and more integral to many aspects of scientific research, computational reproducibility has become an issue of increasing importance to computer systems researchers and domain scientists alike. Though computational reproducibility seems more straight forward than replicating physical experiments, the complex and rapidly changing nature of computer environments makes being able to reproduce and extend such work a serious challenge. In this paper, I explore common reasons that code developed for one research project cannot be successfully executed or extended by subsequent researchers. I review current approaches to these issues, including virtual machines and workflow systems, and their limitations. I then examine how the popular emerging technology Docker combines several areas from systems research - such as operating system virtualization, cross-platform portability, modular re-usable elements, versioning, and a `DevOps' philosophy, to address these challenges. I illustrate this with several examples of Docker use with a focus on the R statistical environment.

Large language models (LLMs) are increasingly deployed in settings where reasoning, such as multi-step problem solving and chain-of-thought, is essential. Yet, current evaluation practices overwhelmingly report single-run accuracy while ignoring the intrinsic uncertainty that naturally arises from stochastic decoding. This omission creates a blind spot because practitioners cannot reliably assess whether a method's reported performance is stable, reproducible, or cost-consistent. We introduce ReasonBENCH, the first benchmark designed to quantify the underlying instability in LLM reasoning. ReasonBENCH provides (i) a modular evaluation library that standardizes reasoning frameworks, models, and tasks, (ii) a multi-run protocol that reports statistically reliable metrics for both quality and cost, and (iii) a public leaderboard to encourage variance-aware reporting. Across tasks from different domains, we find that the vast majority of reasoning strategies and models exhibit high instability. Notably, even strategies with similar average performance can display confidence intervals up to four times wider, and the top-performing methods often incur higher and less stable costs. Such instability compromises reproducibility across runs and, consequently, the reliability of reported performance. To better understand these dynamics, we further analyze the impact of prompts, model families, and scale on the trade-off between solve rate and stability. Our results highlight reproducibility as a critical dimension for reliable LLM reasoning and provide a foundation for future reasoning methods and uncertainty quantification techniques. ReasonBENCH is publicly available at https://github.com/au-clan/ReasonBench .

Deep learning has recently achieved initial success in program analysis tasks such as bug detection. Lacking real bugs, most existing works construct training and test data by injecting synthetic bugs into correct programs. Despite achieving high test accuracy (e.g., 90%), the resulting bug detectors are found to be surprisingly unusable in practice, i.e., <10% precision when used to scan real software repositories. In this work, we argue that this massive performance difference is caused by a distribution shift, i.e., a fundamental mismatch between the real bug distribution and the synthetic bug distribution used to train and evaluate the detectors. To address this key challenge, we propose to train a bug detector in two phases, first on a synthetic bug distribution to adapt the model to the bug detection domain, and then on a real bug distribution to drive the model towards the real distribution. During these two phases, we leverage a multi-task hierarchy, focal loss, and contrastive learning to further boost performance. We evaluate our approach extensively on three widely studied bug types, for which we construct new datasets carefully designed to capture the real bug distribution. The results demonstrate that our approach is practically effective and successfully mitigates the distribution shift: our learned detectors are highly performant on both our test set and the latest version of open source repositories. Our code, datasets, and models are publicly available at https://github.com/eth-sri/learning-real-bug-detector.

Bug reports provide critical insights into software quality, yet existing datasets often suffer from limited scope, outdated content, or insufficient metadata for machine learning. To address these limitations, we present GitBugs-a comprehensive and up-to-date dataset comprising over 150,000 bug reports from nine actively maintained open-source projects, including Firefox, Cassandra, and VS Code. GitBugs aggregates data from Github, Bugzilla and Jira issue trackers, offering standardized categorical fields for classification tasks and predefined train/test splits for duplicate bug detection. In addition, it includes exploratory analysis notebooks and detailed project-level statistics, such as duplicate rates and resolution times. GitBugs supports various software engineering research tasks, including duplicate detection, retrieval augmented generation, resolution prediction, automated triaging, and temporal analysis. The openly licensed dataset provides a valuable cross-project resource for benchmarking and advancing automated bug report analysis. Access the data and code at https://github.com/av9ash/gitbugs/.

Large language models (LLMs) and LLM-based Agents have been applied to fix bugs automatically, demonstrating the capability in addressing software defects by engaging in development environment interaction, iterative validation and code modification. However, systematic analysis of these agent and non-agent systems remain limited, particularly regarding performance variations among top-performing ones. In this paper, we examine seven proprietary and open-source systems on the SWE-bench Lite benchmark for automated bug fixing. We first assess each system's overall performance, noting instances solvable by all or none of these sytems, and explore why some instances are uniquely solved by specific system types. We also compare fault localization accuracy at file and line levels and evaluate bug reproduction capabilities, identifying instances solvable only through dynamic reproduction. Through analysis, we concluded that further optimization is needed in both the LLM itself and the design of Agentic flow to improve the effectiveness of the Agent in bug fixing.

Yes. Can it cure doom loops? Probably not. The Gemma 4 instruction-tuned models share a reproducible failure: on long factual enumeration prompts, such as listing every episode of a TV series, the 88 IAU constellations, or the 151 original Pokemon, they collapse into repetition, either a tight verbatim loop or a list whose entries decay onto a single answer. These loops occur at rates as high as 95% and survive prompt rewording, inference-engine changes, and most sampling adjustments. In this paper we explore whether this behavior is localized enough to remove by weight edits. To localize the cause, we use per-layer ablation and per-neuron attribution, then confirm the strongest candidates with full-generation sweeps. The loops trace to a small set of MLP neurons (or, in the 26B-A4B Mixture-of-Experts model, a few routed experts) which we suppress with static weight edits. These "surgeries" can be as small as a single sign-inverted neuron (in the E2B model). The size of the effective edits grows with model scale, but in all cases, the loop patterns can be addressed at normal generation budgets while preserving general-purpose benchmark scores. However, the edits do not solve everything: we also study longer thinking budgets, where the two larger models most visibly enter doom looping, i.e. a non-convergent regime in which the model self-corrects in circles over a fact it cannot recall, exhausting the budget without committing to a final answer. We show this residual failure is reduced but not eliminated by the same edits, and argue it is fundamentally a knowledge-precision problem rather than a removable circuit; weight surgery can delete a loop, but it cannot supply a missing fact. Our results are both a feasibility demonstration, that is, evidence that a concrete generation pathology can be localized to a few parameters and edited out, and a delineation of where that approach stops.

We present the Consilium Protocol, a Byzantine Fault Tolerance-derived architecture for structured multi-model AI deliberation that treats inter-model disagreement as epistemic signal rather than error. The protocol assigns engineered cognitive personas to language models -- separating what a model is from how it reasons -- and introduces an In-Sample/Out-of-Sample validation framework adapted from quantitative finance to distinguish training-data consensus from empirically grounded conclusions. Across 1,478 deliberation sessions spanning 32 topics in 10 domain categories, we demonstrate that (1) the cognitive persona, not the underlying model, determines epistemic behavior: free edge-inference models costing 0.0002 USD per batch produced comparable analytical output to frontier models costing 10.69 USD; (2) RLHF alignment training creates measurable, domain-specific epistemic blind spots -- contested policy topics exhibit 12.3 percentage points less adversarial challenge than settled science topics, and AI safety topics show asymmetric bias (Δ=11.6%) where models challenge claims that AI is dangerous far more vigorously than claims that AI risk is overstated; (3) the protocol exhibits no directional bias of its own (immigration Δ=2.3%, renewables Δ=1.2%); and (4) out-of-sample evidence retrieval validated 239 claims with 100% evidence retrieval and surfaced 167 blind-spot discoveries invisible to training-data deliberation. Run-to-run reproducibility across randomized modeltimespersona assignments averages pm2.2% standard deviation. Total cost for the complete battery including all overhead: 217 USD. We release the protocol specification under MIT license to enable independent verification.

Bug reports are vital for software maintenance that allow users to inform developers of the problems encountered while using the software. As such, researchers have committed considerable resources toward automating bug replay to expedite the process of software maintenance. Nonetheless, the success of current automated approaches is largely dictated by the characteristics and quality of bug reports, as they are constrained by the limitations of manually-crafted patterns and pre-defined vocabulary lists. Inspired by the success of Large Language Models (LLMs) in natural language understanding, we propose AdbGPT, a new lightweight approach to automatically reproduce the bugs from bug reports through prompt engineering, without any training and hard-coding effort. AdbGPT leverages few-shot learning and chain-of-thought reasoning to elicit human knowledge and logical reasoning from LLMs to accomplish the bug replay in a manner similar to a developer. Our evaluations demonstrate the effectiveness and efficiency of our AdbGPT to reproduce 81.3% of bug reports in 253.6 seconds, outperforming the state-of-the-art baselines and ablation studies. We also conduct a small-scale user study to confirm the usefulness of AdbGPT in enhancing developers' bug replay capabilities.

During Automated Program Repair (APR), it can be challenging to synthesize correct patches for real-world systems in general-purpose programming languages. Recent Large Language Models (LLMs) have been shown to be helpful "copilots" in assisting developers with various coding tasks, and have also been directly applied for patch synthesis. However, most LLMs treat programs as sequences of tokens, meaning that they are ignorant of the underlying semantics constraints of the target programming language. This results in plenty of statically invalid generated patches, impeding the practicality of the technique. Therefore, we propose Repilot, a framework to further copilot the AI "copilots" (i.e., LLMs) by synthesizing more valid patches during the repair process. Our key insight is that many LLMs produce outputs autoregressively (i.e., token by token), resembling human writing programs, which can be significantly boosted and guided through a Completion Engine. Repilot synergistically synthesizes a candidate patch through the interaction between an LLM and a Completion Engine, which 1) prunes away infeasible tokens suggested by the LLM and 2) proactively completes the token based on the suggestions provided by the Completion Engine. Our evaluation on a subset of the widely-used Defects4j 1.2 and 2.0 datasets shows that Repilot fixes 66 and 50 bugs, respectively, surpassing the best-performing baseline by 14 and 16 bugs fixed. More importantly, Repilot is capable of producing more valid and correct patches than the base LLM when given the same generation budget.

Bug reports serve as task specifications for repository-level automated program repair (APR) agents, but they often describe only the observed failure and omit repair-relevant information such as the failure-inducing behavior, behavioral requirement, and implementation scope. As a result, a repair agent may inspect irrelevant code, infer an incorrect requirement, or generate a patch that addresses the reported symptom without restoring the intended repository behavior. We present TrajSpec, a trajectory-guided approach for repository-supported bug report specification refinement. Given an original report and a pre-fix repository, TrajSpec runs a trajectory-collection agent and uses the resulting unverified trajectory as a source of trajectory-derived specification evidence. It organizes this evidence into a three-level representation consisting of a high-level interpretation of the issue, diagnostic findings supporting that interpretation, and concrete repository observations. TrajSpec then generates a draft refined report and applies repository-based review to remove unsupported claims, revise uncertain claims, and add repository-supported details. We evaluate TrajSpec on all 300 SWE-Bench Lite instances using Mini-SWE-Agent V2. TrajSpec's refined reports improve Pass@1 from 41.00% to 59.67% with GPT-5-mini and from 54.67% to 64.33% with MiniMax M2.5. On a stratified sample of 100 instances, TrajSpec's refined reports also improve Pass@1 from 41.00% to 71.00% with Agentless and from 47.00% to 72.00% with AutoCodeRover. Ablation results show that removing repository-based review or the hierarchical evidence representation reduces Pass@1 from 59.67% to 48.00% and 47.67%, respectively. Overall, TrajSpec provides actionable repository-supported context that consistently improves repair performance.

Recent mixed-policy optimization methods for LLM reasoning that interleave or blend supervised and reinforcement learning signals report improvements over the standard SFT-then-RL pipeline. We show that numerous recently published research papers rely on a faulty baseline caused by two distinct bugs: a CPU-offloaded optimizer bug in DeepSpeed that silently drops intermediate micro-batches during gradient accumulation (affecting multiple downstream frameworks including TRL, OpenRLHF and Llama-Factory), and a loss aggregation bug in OpenRLHF that incorrectly weights per-mini-batch losses. Together they suppress SFT performance, with the optimizer bug accounting for most of the gap and the loss aggregation bug contributing a smaller additional effect. Once corrected, the standard SFT-then-RL pipeline surpasses every published mixed-policy method we evaluate by +3.8 points on math benchmarks with Qwen2.5-Math-7B and by +22.2 points with Llama-3.1-8B. Even a truncated variant with just 50 RL steps outperforms mixed-policy methods on math benchmarks while using fewer FLOPs.

Autonomous research systems increasingly make the scientific workflow executable: agents can propose ideas, run code, inspect results, and draft papers. But executable workflows do not by themselves produce research judgment. We analyze where current systems lose trial experience: weak evidence becomes prose, pilot signals become broad claims, memory remains textual, and recurring process failures do not change later behavior. We introduce Sibyl-AutoResearch, a self-evolving AutoResearch framework built around Scientific Trial-and-Error Harnesses. A harness lets agents run bounded trials, preserve positive and negative outcomes, and route lessons into later planning, validation, claim scope, scheduling, critique, writing, and harness repair. We formalize this through two auditable conversion units: trial-to-behavior conversion, which links trial signals to later research actions, and trial-to-harness-behavior conversion, which links recurring process failures to system updates. We implement the framework in SIBYL, a file-backed autonomous research system that exposes the state, roles, memory, gates, and artifact traces needed to inspect these conversion paths. A retrospective audit identifies eight high-confidence conversion events, with a median latency of one iteration and a maximum latency of three iterations. A recovered-failure registry further shows how five naturally occurring failure classes, including duplicate results, stale numbers, and unsupported statistics, were blocked, downgraded, or routed into later repair. These traces do not establish a comparative performance claim; they show that the proposed conversion units are recoverable from realistic autonomous-research workspaces. The SIBYL framework and system are available at https://github.com/Sibyl-Research-Team/AutoResearch-SibylSystem.

We study whether demonstration-curation metrics that detect defective training episodes also improve the downstream behavior-cloning policy that trains on the curated data. On a contact-rich LIBERO pick-and-place benchmark with a controlled structural defect (early gripper release during the carry phase), we find that the two quantities are sharply decoupled. The metric with the highest defect-detection AUROC (0.804) produces the worst curated policy (13.3% task success), while a metric with a substantially lower AUROC (0.638) produces a policy that nearly matches the oracle trained on ground-truth clean data (90.0% vs. 93.3%). We further show that five of the seven metrics we evaluate exploit episode length as a trivial proxy for the defect label, a confound that inflates reported AUROCs to near-perfect values and disappears once episode length is controlled. Across all conditions, the contaminated baseline succeeds on only 3.3% of rollouts, and the two best curation methods close this to within 3 percentage points of the 93.3% oracle ceiling. Our results argue that curation methods should be evaluated by the policy they produce, not the defects they flag, and that any curation benchmark must control for episode length before reporting detection accuracy. We release the testbed, all metric implementations, and the evaluation pipeline.

Large language models of code (Code-LLMs) have recently brought tremendous advances to code completion, a fundamental feature of programming assistance and code intelligence. However, most existing works ignore the possible presence of bugs in the code context for generation, which are inevitable in software development. Therefore, we introduce and study the buggy-code completion problem, inspired by the realistic scenario of real-time code suggestion where the code context contains potential bugs -- anti-patterns that can become bugs in the completed program. To systematically study the task, we introduce two datasets: one with synthetic bugs derived from semantics-altering operator changes (buggy-HumanEval) and one with realistic bugs derived from user submissions to coding problems (buggy-FixEval). We find that the presence of potential bugs significantly degrades the generation performance of the high-performing Code-LLMs. For instance, the passing rates of CodeGen-2B-mono on test cases of buggy-HumanEval drop more than 50% given a single potential bug in the context. Finally, we investigate several post-hoc methods for mitigating the adverse effect of potential bugs and find that there remains a large gap in post-mitigation performance.

Large Language Models (LLMs) have demonstrated strong natural language processing and code synthesis capabilities, which has led to their rapid adoption in software engineering applications. However, details about LLM training data are often not made public, which has caused concern as to whether existing bug benchmarks are included. In lieu of the training data for the popular GPT models, we examine the training data of the open-source LLM StarCoder, and find it likely that data from the widely used Defects4J benchmark was included, raising the possibility of its inclusion in GPT training data as well. This makes it difficult to tell how well LLM-based results on Defects4J would generalize, as for any results it would be unclear whether a technique's performance is due to LLM generalization or memorization. To remedy this issue and facilitate continued research on LLM-based SE, we present the GitHub Recent Bugs (GHRB) dataset, which includes 76 real-world Java bugs that were gathered after the OpenAI data cut-off point.

Anomaly detection methods identify examples that do not follow the expected behaviour, typically in an unsupervised fashion, by assigning real-valued anomaly scores to the examples based on various heuristics. These scores need to be transformed into actual predictions by thresholding, so that the proportion of examples marked as anomalies equals the expected proportion of anomalies, called contamination factor. Unfortunately, there are no good methods for estimating the contamination factor itself. We address this need from a Bayesian perspective, introducing a method for estimating the posterior distribution of the contamination factor of a given unlabeled dataset. We leverage on outputs of several anomaly detectors as a representation that already captures the basic notion of anomalousness and estimate the contamination using a specific mixture formulation. Empirically on 22 datasets, we show that the estimated distribution is well-calibrated and that setting the threshold using the posterior mean improves the anomaly detectors' performance over several alternative methods. All code is publicly available for full reproducibility.

The foundation of reproducible science lies in protocols that are precise, logically ordered, and executable. The autonomous generation of these protocols through natural language queries could greatly improve the efficiency of the reproduction process. However, current leading large language models (LLMs) often generate incomplete or inconsistent protocols, limiting their utility. To address this limitation, we first introduce SciRecipe, a large-scale dataset of over 12K structured protocols spanning 27 biological subfields and encompassing both comprehension and problem-solving tasks. To further improve protocol generation, we propose the "Sketch-and-Fill" paradigm, which separates analysis, structuring, and expression to ensure each step is explicit and verifiable. Complementing this, the structured component-based reward mechanism evaluates step granularity, action order, and semantic fidelity, aligning model optimization with experimental reliability. Building on these components, we develop Thoth, trained through a staged Knowledge-to-Action process that progresses from knowledge acquisition to operational reasoning and ultimately to robust, executable protocol generation. Across multiple benchmarks, Thoth consistently surpasses both proprietary and open-source LLMs, achieving significant improvements in step alignment, logical sequencing, and semantic accuracy. Our approach paves the way for reliable scientific assistants that bridge knowledge with experimental execution. All data, code, and models will be released publicly.

Imitation-learning policies inherit the quality of the demonstrations they are trained on, and a growing set of curation metrics promise to score and filter low-quality demonstrations automatically. These metrics are each validated on different data with different protocols, so it is unclear which of them actually identify the demonstrations that harm a policy. We build a controlled testbed in which demonstration defects are injected with known type, and audit seven curation metrics along two axes: how well each separates defective from clean demonstrations, and whether training a behavior-cloning policy on each metric's curated subset improves task success. We study two defect regimes. Subtle perturbations (correlated action noise, tremor, truncation) are detectable by multivariate outlier scoring and, once removed, recover the full downstream gap. Structural errors, where the demonstration executes a wrong action at a key moment, are invisible to every action-only metric we test, and two of them are inverted: they score defective demonstrations as higher quality and, used for curation, tend to leave the policy at or below the uncurated baseline rather than above it. Only metrics that examine the state trajectory detect structural errors, and even the best of them recovers just a third of the downstream gap. High detection accuracy does not guarantee downstream improvement. We release the testbed and all curation implementations.

Large Language Models (LLMs) are now integral across various domains and have demonstrated impressive performance. Progress, however, rests on the premise that benchmark scores are both accurate and reproducible. We demonstrate that the reproducibility of LLM performance is fragile: changing system configuration such as evaluation batch size, GPU count, and GPU version can introduce significant difference in the generated responses. This issue is especially pronounced in reasoning models, where minor rounding differences in early tokens can cascade into divergent chains of thought, ultimately affecting accuracy. For instance, under bfloat16 precision with greedy decoding, a reasoning model like DeepSeek-R1-Distill-Qwen-7B can exhibit up to 9% variation in accuracy and 9,000 tokens difference in response length due to differences in GPU count, type, and evaluation batch size. We trace the root cause of this variability to the non-associative nature of floating-point arithmetic under limited numerical precision. This work presents the first systematic investigation into how numerical precision affects reproducibility in LLM inference. Through carefully controlled experiments across various hardware, software, and precision settings, we quantify when and how model outputs diverge. Our analysis reveals that floating-point precision -- while critical for reproducibility -- is often neglected in evaluation practices. Inspired by this, we develop a lightweight inference pipeline, dubbed LayerCast, that stores weights in 16-bit precision but performs all computations in FP32, balancing memory efficiency with numerical stability. Code is available at https://github.com/nanomaoli/llm_reproducibility.

In modern data-driven science, reproducibility and reusability are key challenges. Scientists are well skilled in the process from data to publication. Although some publication channels require source code and data to be made accessible, rerunning and verifying experiments is usually hard due to a lack of standards. Therefore, reusing existing scientific data processing code from state-of-the-art research is hard as well. This is why we introduce CLAIMED, which has a proven track record in scientific research for addressing the repeatability and reusability issues in modern data-driven science. CLAIMED is a framework to build reusable operators and scalable scientific workflows by supporting the scientist to draw from previous work by re-composing workflows from existing libraries of coarse-grained scientific operators. Although various implementations exist, CLAIMED is programming language, scientific library, and execution environment agnostic.

The increasing development of large language models (LLMs) in code generation has drawn significant attention among researchers. To enhance LLM-based code generation ability, current efforts are predominantly directed towards collecting high-quality datasets and leveraging diverse training technologies. However, there is a notable lack of comprehensive studies examining the limitations and boundaries of these existing methods. To bridge this gap, we conducted an extensive empirical study evaluating the performance of three leading closed-source LLMs and four popular open-source LLMs on three commonly used benchmarks. Our investigation, which evaluated the length, cyclomatic complexity and API number of the generated code, revealed that these LLMs face challenges in generating successful code for more complex problems, and tend to produce code that is shorter yet more complicated as compared to canonical solutions. Additionally, we developed a taxonomy of bugs for incorrect codes that includes three categories and 12 sub-categories, and analyze the root cause for common bug types. Furthermore, to better understand the performance of LLMs in real-world projects, we manually created a real-world benchmark comprising 140 code generation tasks. Our analysis highlights distinct differences in bug distributions between actual scenarios and existing benchmarks. Finally, we propose a novel training-free iterative method that introduces self-critique, enabling LLMs to critique and correct their generated code based on bug types and compiler feedback. Experimental results demonstrate that our approach can significantly mitigate bugs and increase the passing rate by 29.2% after two iterations, indicating substantial potential for LLMs to handle more complex problems.

Consistently checking the statistical significance of experimental results is one of the mandatory methodological steps to address the so-called "reproducibility crisis" in deep reinforcement learning. In this tutorial paper, we explain how the number of random seeds relates to the probabilities of statistical errors. For both the t-test and the bootstrap confidence interval test, we recall theoretical guidelines to determine the number of random seeds one should use to provide a statistically significant comparison of the performance of two algorithms. Finally, we discuss the influence of deviations from the assumptions usually made by statistical tests. We show that they can lead to inaccurate evaluations of statistical errors and provide guidelines to counter these negative effects. We make our code available to perform the tests.

Mobile GUI agents powered by large language models have progressed rapidly, creating urgent needs for realistic and comprehensive evaluation. Existing benchmarks prioritize reproducibility but are often limited to open-source apps or file-operation tasks for the difficulty of constructing rewards on real applications, leaving a gap between benchmark settings and real-world usage. Moreover, most benchmarks focus on basic grounding and navigation, with limited coverage of complex, long-horizon interactions. To address these limitations, we introduce SimuWoB, a fully synthetic benchmark for mobile GUI agents with 120 challenging tasks spanning diverse types and difficulty levels. We build a robust virtual environment generation framework that synthesizes high-fidelity tasks and environments, and automatically provides valid rewards for each task. Each environment is deployed as a backend-free webpage accessible via URL, enabling efficient and reproducible evaluation. We conduct comprehensive experiments on several state-of-the-art mobile GUI agents. The average success rate is only 27.92%, dropping to 17.82% on long-horizon tasks, which reveals substantial weaknesses in current agents under complex scenarios. Evaluation result comparison with real-world sample tasks demonstrate that agent assessments based on our synthetic environment generalize well. We further provide diagnostic insights across key capability dimensions and discuss implications for future mobile GUI agent development.

Bug bounty platforms (e.g., HackerOne, BugCrowd) leverage crowd-sourced vulnerability discovery to improve continuous coverage, reduce the cost of discovery, and serve as an integral complement to internal red teams. With the rise of AI-generated bug reports, little work exists to help bug hunters understand why these reports are labeled as invalid. To improve report quality and reduce reviewers' burden, it is critical to predict invalid reports and interpret invalid reasons. In this work, we conduct an empirical study with the purpose of helping bug hunters understand the validity of reports. We collect a dataset of 9,942 disclosed bug bounty reports, including 1,400 invalid reports, and evaluate whether state-of-the-art large language models can identify invalid reports. While models such as GPT-5, DeepSeek, and a fine-tuned RoBERTa achieve strong overall accuracy, they consistently struggle to detect invalid cases, showing a tendency to over-accept reports. To improve invalidity detection, we build a taxonomy of rejection reasons for Information Disclosure vulnerabilities and incorporate it into a retrieval-augmented generation (RAG) framework. This approach substantially improves classification consistency and reduces bias. We also examine whether reviewer decisions may be influenced by factors beyond the content of the report. Our analysis shows that reporters with higher reputations tend to receive more favorable outcomes in borderline cases, suggesting that perceived expertise can influence review judgments. Overall, our findings highlight the challenges of invalid report identification and show that combining LLMs with structured reviewer knowledge can support more transparent and consistent vulnerability report review.

Turning a research idea into a complete paper requires more than text generation: the system must retrieve literature, design and execute experiments, revise claims according to evidence, produce publication-ready figures, and maintain consistency across a long generation process. We present Spark-to-Paper, an end-to-end research paper generation system implemented as thirteen composable skills inside an existing coding assistant, without requiring a separate agent platform or orchestration service. Spark-to-Paper separates model-based judgment from deterministic operations that can be directly executed and checked. It further separates experiment planning from reporting, so that required evidence is specified before results are observed and manuscript claims are revised according to measured outcomes. To improve reliability over long research trajectories, the system combines deterministic integrity checks with self-critique and bounds a failure mode we call the Self-Refutation Loop, in which repeated experiments continue to reject the original research objective. Spark-to-Paper also produces editable vector figures through programmatic plotting for experimental results and code-based reconstruction for generated method diagrams. Across eight controlled research topics, Spark-to-Paper achieves 99.5% citation validity and 96.4% figure editability. A controlled ablation increases fabrication detection from 14% for a single-pass draft to 92% with the full integrity and review stack, while adversarial review achieves 74% precision. The full system uses 11.9M tokens, costs $8.1 per manuscript, and requires 3.2 hours on average. These results show that end-to-end research paper generation can be implemented as a lightweight, composable workflow inside existing coding assistants while keeping experimental evidence central to how claims are accepted, revised, or abandoned.

This paper presents PyResBugs, a curated dataset of residual bugs, i.e., defects that persist undetected during traditional testing but later surface in production, collected from major Python frameworks. Each bug in the dataset is paired with its corresponding fault-free (fixed) version and annotated with multi-level natural language (NL) descriptions. These NL descriptions enable natural language-driven fault injection, offering a novel approach to simulating real-world faults in software systems. By bridging the gap between software fault injection techniques and real-world representativeness, PyResBugs provides researchers with a high-quality resource for advancing AI-driven automated testing in Python systems.

Tool-augmented large language models have advanced from single-turn question answering to deep research workflows that iteratively plan queries, invoke external tools, and synthesize information to address complex information needs. Evaluating such workflows presents a fundamental challenge: reliance on live APIs introduces non-determinism, as tool invocations may yield different results across runs due to temporal drift, rate limiting, and evolving backend states. This variance undermines reproducibility and invalidates cross-system comparisons. We present ScholarGym, a simulation environment for reproducible evaluation of deep research workflows on academic literature. The environment decouples workflow components into query planning, tool invocation, and relevance assessment, enabling fine-grained analysis of each stage under controlled conditions. Built on a static corpus of 570K papers with deterministic retrieval, ScholarGym provides 2,536 queries with expert-annotated ground truth. Experiments across diverse backbone models reveal how reasoning capabilities, planning strategies, and selection mechanisms interact over iterative refinement.

We introduce a new -- currently 42 gigabyte -- ``living'' dataset of phone images of dog feces, annotated with manually drawn or AI-assisted polygon labels. There are 6k full resolution images and 4k detailed polygon annotations. The collection and annotation of images started in late 2020 and the dataset grows by roughly 1GB a month. We train VIT and MaskRCNN baseline models to explore the difficulty of the dataset. The best model achieves a pixelwise average precision of 0.858 on a 691-image validation set and 0.847 on a small independently captured 30-image contributor test set. The most recent snapshot of dataset is made publicly available through three different distribution methods: one centralized (Girder) and two decentralized (IPFS and BitTorrent). We study of the trade-offs between distribution methods and discuss the feasibility of each with respect to reliably sharing open scientific data. The code to reproduce the experiments is hosted on GitHub, and the data is published under the Creative Commons Attribution 4.0 International license. Model weights are made publicly available with the dataset. Experimental hardware, time, energy, and emissions are quantified.

Bug reports contain the information developers need to triage and fix software bugs. However, unclear, incomplete, or ambiguous information may lead to delays and excessive manual effort spent on bug triage and resolution. In this paper, we explore whether Instruction fine-tuned Large Language Models (LLMs) can automatically transform casual, unstructured bug reports into high-quality, structured bug reports adhering to a standard template. We evaluate three open-source instruction-tuned LLMs (Qwen 2.5, Mistral, and Llama 3.2) against ChatGPT-4o, measuring performance on established metrics such as CTQRS, ROUGE, METEOR, and SBERT. Our experiments show that fine-tuned Qwen 2.5 achieves a CTQRS score of 77%, outperforming both fine-tuned Mistral (71%), Llama 3.2 (63%) and ChatGPT in 3-shot learning (75%). Further analysis reveals that Llama 3.2 shows higher accuracy of detecting missing fields particularly Expected Behavior and Actual Behavior, while Qwen 2.5 demonstrates superior performance in capturing Steps-to-Reproduce, with an F1 score of 76%. Additional testing of the models on other popular projects (e.g., Eclipse, GCC) demonstrates that our approach generalizes well, achieving up to 70% CTQRS in unseen projects' bug reports. These findings highlight the potential of instruction fine-tuning in automating structured bug report generation, reducing manual effort for developers and streamlining the software maintenance process.

The autonomous discovery of bugs remains a significant challenge in modern software development. Compared to code generation, the complexity of dynamic runtime environments makes bug discovery considerably harder for large language models (LLMs). In this paper, we take game development as a representative domain and introduce the Game Benchmark for Quality Assurance (GBQA), a benchmark containing 30 games and 124 human-verified bugs across three difficulty levels, to evaluate whether LLMs can autonomously detect software bugs. The benchmark is constructed using a multi-agent system that develops games and injects bugs in a scalable manner, with human experts in the loop to ensure correctness. Moreover, we provide a baseline interactive agent equipped with a multi-round ReAct loop and a memory mechanism, enabling long-horizon exploration of game environments for bug detection across different LLMs. Extensive experiments on frontier LLMs demonstrate that autonomous bug discovery remains highly challenging: the best-performing model, Claude-4.6-Opus in thinking mode, identifies only 48.39% of the verified bugs. We believe GBQA provides an adequate testbed and evaluation criterion, and that further progress on it will help close the gap in autonomous software engineering.

How do multi-turn reasoning systems fail? The expected answer is logical contradiction, in which the system's maintained state becomes unsatisfiable. We show that the dominant mode is instead satisfiable drift, where the internal state stays consistent while the returned answer silently violates prior commitments. We build DRIFT-Bench (Decomposing Reasoning Into Failure Types), a solver-instrumented benchmark of 816 test problems across three constraint domains, and evaluate four methods on it across four open-weight models (8B-120B parameters). MUS-Repair, which feeds minimal unsatisfiable subsets back to the generator, is strongest in every setting (+1.8 to +15.0 pp over the best non-MUS baseline). But the central finding is what repair leaves behind. After structured feedback, models rarely contradict themselves. They forget. Residual errors are 98-100% satisfiable drift across all settings, while contradiction drops to near zero. Reliable multi-turn systems must separately validate that the returned answer respects the maintained state. Code is available at https://github.com/kaons-research/drift-bench.

Despite recent advances in inversion and instruction-based image editing, existing approaches primarily excel at editing single, prominent objects but significantly struggle when applied to complex scenes containing multiple entities. To quantify this gap, we first introduce RefEdit-Bench, a rigorous real-world benchmark rooted in RefCOCO, where even baselines trained on millions of samples perform poorly. To overcome this limitation, we introduce RefEdit -- an instruction-based editing model trained on our scalable synthetic data generation pipeline. Our RefEdit, trained on only 20,000 editing triplets, outperforms the Flux/SD3 model-based baselines trained on millions of data. Extensive evaluations across various benchmarks demonstrate that our model not only excels in referring expression tasks but also enhances performance on traditional benchmarks, achieving state-of-the-art results comparable to closed-source methods. We release data \& checkpoint for reproducibility.

Evaluating the quality of synthetic data remains a key challenge for ensuring privacy and utility in data-driven research. In this work, we present an evaluation framework that quantifies how well synthetic data replicates original distributional properties while ensuring privacy. The proposed approach employs a holdout-based benchmarking strategy that facilitates quantitative assessment through low- and high-dimensional distribution comparisons, embedding-based similarity measures, and nearest-neighbor distance metrics. The framework supports various data types and structures, including sequential and contextual information, and enables interpretable quality diagnostics through a set of standardized metrics. These contributions aim to support reproducibility and methodological consistency in benchmarking of synthetic data generation techniques. The code of the framework is available at https://github.com/mostly-ai/mostlyai-qa.

The exercise of detecting similar bug reports in bug tracking systems is known as duplicate bug report detection. Having prior knowledge of a bug report's existence reduces efforts put into debugging problems and identifying the root cause. Rule and Query-based solutions recommend a long list of potential similar bug reports with no clear ranking. In addition, triage engineers are less motivated to spend time going through an extensive list. Consequently, this deters the use of duplicate bug report retrieval solutions. In this paper, we have proposed a solution using a combination of NLP techniques. Our approach considers unstructured and structured attributes of a bug report like summary, description and severity, impacted products, platforms, categories, etc. It uses a custom data transformer, a deep neural network, and a non-generalizing machine learning method to retrieve existing identical bug reports. We have performed numerous experiments with significant data sources containing thousands of bug reports and showcased that the proposed solution achieves a high retrieval accuracy of 70% for recall@5.

Lack of rigorous reproducibility and validation are major hurdles for scientific development across many fields. Materials science in particular encompasses a variety of experimental and theoretical approaches that require careful benchmarking. Leaderboard efforts have been developed previously to mitigate these issues. However, a comprehensive comparison and benchmarking on an integrated platform with multiple data modalities with both perfect and defect materials data is still lacking. This work introduces JARVIS-Leaderboard, an open-source and community-driven platform that facilitates benchmarking and enhances reproducibility. The platform allows users to set up benchmarks with custom tasks and enables contributions in the form of dataset, code, and meta-data submissions. We cover the following materials design categories: Artificial Intelligence (AI), Electronic Structure (ES), Force-fields (FF), Quantum Computation (QC) and Experiments (EXP). For AI, we cover several types of input data, including atomic structures, atomistic images, spectra, and text. For ES, we consider multiple ES approaches, software packages, pseudopotentials, materials, and properties, comparing results to experiment. For FF, we compare multiple approaches for material property predictions. For QC, we benchmark Hamiltonian simulations using various quantum algorithms and circuits. Finally, for experiments, we use the inter-laboratory approach to establish benchmarks. There are 1281 contributions to 274 benchmarks using 152 methods with more than 8 million data-points, and the leaderboard is continuously expanding. The JARVIS-Leaderboard is available at the website: https://pages.nist.gov/jarvis_leaderboard

Test-input generation for tensor kernels is folkloric. Most projects pick a representative shape and dtype, run a fixed-shape allclose-style check, and ship. We make the choices explicit and measure them. Using the gpuemu op-schema-aware seeded fuzzer (arXiv:2606.20128), we evaluate seven test-generation strategies across a 26-op corpus (16 correct controls and 10 LLM-style buggy variants seeded with documented transcription patterns) on an RTX 3060 GPU instance. Strategies vary the shape candidate set, the dtype mix, and the input value distribution. We report each strategy on two axes: bug recall and control false-positive (FP) rate. Boundary-only shape sampling is the operationally safe winner: 78% recall on the 10 buggy kernels with 0% FP on the 16 controls. Adversarial value sampling reaches higher recall (99%) but inflates control FP to 94% because the strategy injects NaN and Inf inputs and the validator's NaN check fires on every kernel that propagates them, not only on buggy kernels. On the two softmax tail-mask bugs the "regular" strategy (no boundary shapes) catches 0%, while boundary raises recall to 100% and 62% respectively. That gap is the clearest single signal in the data. The corpus result is about which seeded bug patterns each strategy catches, not about the bug rate of any specific deployed LLM.

Accelerating scientific discovery requires the identification of which experiments would yield the best outcomes before committing resources to costly physical validation. While existing benchmarks evaluate LLMs on scientific knowledge and reasoning, their ability to predict experimental outcomes - a task where AI could significantly exceed human capabilities - remains largely underexplored. We introduce SciPredict, a benchmark comprising 405 tasks derived from recent empirical studies in 33 specialized sub-fields of physics, biology, and chemistry. SciPredict addresses two critical questions: (a) can LLMs predict the outcome of scientific experiments with sufficient accuracy? and (b) can such predictions be reliably used in the scientific research process? Evaluations reveal fundamental limitations on both fronts. Model accuracies are 14-26% and human expert performance is approx20%. Although some frontier models exceed human performance model accuracy is still far below what would enable reliable experimental guidance. Even within the limited performance, models fail to distinguish reliable predictions from unreliable ones, achieving only approx20% accuracy regardless of their confidence or whether they judge outcomes as predictable without physical experimentation. Human experts, in contrast, demonstrate strong calibration: their accuracy increases from approx5% to approx80% as they deem outcomes more predictable without conducting the experiment. SciPredict establishes a rigorous framework demonstrating that superhuman performance in experimental science requires not just better predictions, but better awareness of prediction reliability. For reproducibility all our data and code are provided at https://github.com/scaleapi/scipredict

Large Language Models (LLMs) are being increasingly used in software engineering tasks, with an increased focus on bug report resolution over the past year. However, most proposed systems fail to properly handle uncertain or incorrect inputs and outputs. Existing LLM-based tools and coding agents respond to every issue and generate a patch for every case, even when the input is vague or their own output is incorrect. There are no mechanisms in place to abstain when confidence is low. This leads to unreliable behaviour, such as hallucinated code changes or responses based on vague issue reports. We introduce BouncerBench, a benchmark that evaluates whether LLM-based software agents can refuse to act when inputs are ill-defined or refuse to respond when their own outputs are likely to be incorrect. Unlike prior benchmarks that implicitly incentivize models to generate responses even when uncertain, BouncerBench aims to improve precision by targeting two overlooked failure points: (1) vague or underspecified issue descriptions in tickets and (2) logically or functionally incorrect code patches created by the system. It measures whether proposed systems can distinguish actionable issues from vague tickets and valid patches from untrustworthy ones. We also implement a basic input and output bouncer, evaluating how well current LLMs can abstain when needed. Our results show that most models fail to abstain from underspecified inputs or incorrect outputs. Hence, we conclude that there is significant room for improvement before LLMs can be trusted to make correct decisions and recommendations in real-world software engineering workflows. BouncerBench provides a first step toward evaluating and building more cautious, trustworthy code agents. The replication package, dataset, and leaderboard can be found at bouncerbench.com

Machine learning-based program analyses have recently shown the promise of integrating formal and probabilistic reasoning towards aiding software development. However, in the absence of large annotated corpora, training these analyses is challenging. Towards addressing this, we present BugLab, an approach for self-supervised learning of bug detection and repair. BugLab co-trains two models: (1) a detector model that learns to detect and repair bugs in code, (2) a selector model that learns to create buggy code for the detector to use as training data. A Python implementation of BugLab improves by up to 30% upon baseline methods on a test dataset of 2374 real-life bugs and finds 19 previously unknown bugs in open-source software.

Jupyter notebooks are widely used for machine learning (ML) prototyping. Yet few debugging tools are designed for ML code in notebooks, potentially due to the lack of benchmarks. We introduce JunoBench, the first benchmark dataset of real-world crashes in Python-based ML notebooks. JunoBench has 111 curated and reproducible crashes from public Kaggle notebooks, each paired with a verifiable fix, ranging over popular ML libraries, including TensorFlow/Keras, PyTorch, Scikit-learn, Pandas, and NumPy, as well as notebook-specific out-of-order execution issue. To support reproducibility and ease of use, JunoBench offers a unified execution environment where crashes and fixes can be reliably reproduced. By providing realistic crashes and their resolutions, JunoBench facilitates bug detection, localization, and repair tailored to the interactive and iterative nature of notebook-based ML development.

Automated issue solving aims to resolve real-world issues in software repositories. The most popular benchmarks for automated issue solving are SWE-bench and its human-filtered subset SWE-bench Verified. These benchmarks leverage testing to validate generated patches. However, because testing is rarely exhaustive, a patch may pass the tests but nevertheless fail to match the developers' expectations. Unfortunately, it is currently unclear to what extent evaluations performed with SWE-bench suffer from such plausible but incorrect patches. This paper presents an in-depth empirical study of the correctness of plausible patches generated by three state-of-the-art issue-solving tools evaluated on SWE-bench Verified. We extensively test and inspect generated patches, and compare them against human-written ground truth patches. The core of our methodology is a novel technique PatchDiff for differential patch testing, which automatically exposes behavioral discrepancies between two patches. Our findings reveal critical weaknesses in SWE-bench's patch validation mechanism, which causes 7.8% of all patches to count as correct while failing the developer-written test suite. Moreover, our novel automated technique reveals that even more (29.6%) plausible patches induce different behavior than the ground truth patches. These behavioral differences are often due to similar, but divergent implementations (46.8%) and due to generated patches that adapt more behavior than the ground truth patches (27.3%). Our manual inspection shows that 28.6% of behaviorally divergent patches are certainly incorrect. Combined, the different weaknesses lead to an inflation of reported resolution rates by 6.2 absolute percent points. Our findings are a call to arms for more robust and reliable evaluation of issue-solving tools. We envision our automated differential patch testing technique to be useful for this purpose.

We present HaPy-Bug, a curated dataset of 793 Python source code commits associated with bug fixes, with each line of code annotated by three domain experts. The annotations offer insights into the purpose of modified files, changes at the line level, and reviewers' confidence levels. We analyze HaPy-Bug to examine the distribution of file purposes, types of modifications, and tangled changes. Additionally, we explore its potential applications in bug tracking, the analysis of bug-fixing practices, and the development of repository analysis tools. HaPy-Bug serves as a valuable resource for advancing research in software maintenance and security.

Reproducibility in scientific work has been becoming increasingly important in research communities such as machine learning, natural language processing, and computer vision communities due to the rapid development of the research domains supported by recent advances in deep learning. In this work, we present a significantly upgraded version of torchdistill, a modular-driven coding-free deep learning framework significantly upgraded from the initial release, which supports only image classification and object detection tasks for reproducible knowledge distillation experiments. To demonstrate that the upgraded framework can support more tasks with third-party libraries, we reproduce the GLUE benchmark results of BERT models using a script based on the upgraded torchdistill, harmonizing with various Hugging Face libraries. All the 27 fine-tuned BERT models and configurations to reproduce the results are published at Hugging Face, and the model weights have already been widely used in research communities. We also reimplement popular small-sized models and new knowledge distillation methods and perform additional experiments for computer vision tasks.

Frontier AI agents show increasing promise as scientific research assistants, and may eventually be useful for extended, open-ended research workflows. However, in order to use agents for novel research, we must first assess the underlying faithfulness and correctness of their work. To evaluate agents as research assistants, we introduce ReplicationBench, an evaluation framework that tests whether agents can replicate entire research papers drawn from the astrophysics literature. Astrophysics, where research relies heavily on archival data and computational study while requiring little real-world experimentation, is a particularly useful testbed for AI agents in scientific research. We split each paper into tasks which require agents to replicate the paper's core contributions, including the experimental setup, derivations, data analysis, and codebase. Each task is co-developed with the original paper authors and targets a key scientific result, enabling objective evaluation of both faithfulness (adherence to original methods) and correctness (technical accuracy of results). ReplicationBench is extremely challenging for current frontier language models: even the best-performing language models score under 20%. We analyze ReplicationBench trajectories in collaboration with domain experts and find a rich, diverse set of failure modes for agents in scientific research. ReplicationBench establishes the first benchmark of paper-scale, expert-validated astrophysics research tasks, reveals insights about agent performance generalizable to other domains of data-driven science, and provides a scalable framework for measuring AI agents' reliability in scientific research.

Reproducibility and replicability are critical pillars of empirical research, particularly in machine learning, where they depend not only on the availability of models, but also on the datasets used to train and evaluate those models. In this paper, we introduce the Construction Industry Steel Ordering List (CISOL) dataset, which was developed with a focus on transparency to ensure reproducibility, replicability, and extensibility. CISOL provides a valuable new research resource and highlights the importance of having diverse datasets, even in niche application domains such as table extraction in civil engineering. CISOL is unique in that it contains real-world civil engineering documents from industry, making it a distinctive contribution to the field. The dataset contains more than 120,000 annotated instances in over 800 document images, positioning it as a medium-sized dataset that provides a robust foundation for Table Structure Recognition (TSR) and Table Detection (TD) tasks. Benchmarking results show that CISOL achieves 67.22 mAP@0.5:0.95:0.05 using the YOLOv8 model, outperforming the TSR-specific TATR model. This highlights the effectiveness of CISOL as a benchmark for advancing TSR, especially in specialized domains.

AI pentesting agents are increasingly credible as offensive security systems, but current benchmarks still provide limited guidance on which will perform best in real-world targets. Existing evaluation protocols assess and optimize for predefined goals such as capture-the-flag, remote code execution, exploit reproduction, or trajectory similarity, in simplified or narrow settings. These tools are valuable for measuring bounded capabilities, yet they do not adequately capture the complexity, open-ended exploration, and strategic decision-making required in realistic pentesting. In this paper, we present a practical evaluation protocol that shifts assessment from task completion to validated vulnerability discovery, allowing evaluation in sufficiently complex targets spanning multiple attack surfaces and vulnerability classes. The protocol combines structured ground-truth with LLM-based semantic matching to identify vulnerabilities, bipartite resolution to score findings under realistic ambiguity, continuous ground-truth maintenance, repeated and cumulative evaluation of stochastic agents, efficiency metrics, and reduced-suite selection for sustainable experimentation. This protocol extends the state of the art by enabling a more realistic, operationally informative comparison of AI pentesting agents. To enable reproducibility, we also release expert-annotated ground truth and code for the proposed evaluation protocol: https://github.com/ethiack/ethibench.

Large Language Models (LLMs) like GPT-4, LLaMA, and Qwen have demonstrated remarkable success across a wide range of applications. However, these models remain inherently vulnerable to prompt injection attacks, which can bypass existing safety mechanisms, highlighting the urgent need for more robust attack detection methods and comprehensive evaluation benchmarks. To address these challenges, we introduce GenTel-Safe, a unified framework that includes a novel prompt injection attack detection method, GenTel-Shield, along with a comprehensive evaluation benchmark, GenTel-Bench, which compromises 84812 prompt injection attacks, spanning 3 major categories and 28 security scenarios. To prove the effectiveness of GenTel-Shield, we evaluate it together with vanilla safety guardrails against the GenTel-Bench dataset. Empirically, GenTel-Shield can achieve state-of-the-art attack detection success rates, which reveals the critical weakness of existing safeguarding techniques against harmful prompts. For reproducibility, we have made the code and benchmarking dataset available on the project page at https://gentellab.github.io/gentel-safe.github.io/.

Large Language Models (LLMs) for code have gained significant attention recently. They can generate code in different programming languages based on provided prompts, fulfilling a long-lasting dream in Software Engineering (SE), i.e., automatic code generation. Similar to human-written code, LLM-generated code is prone to bugs, and these bugs have not yet been thoroughly examined by the community. Given the increasing adoption of LLM-based code generation tools (e.g., GitHub Copilot) in SE activities, it is critical to understand the characteristics of bugs contained in code generated by LLMs. This paper examines a sample of 333 bugs collected from code generated using three leading LLMs (i.e., CodeGen, PanGu-Coder, and Codex) and identifies the following 10 distinctive bug patterns: Misinterpretations, Syntax Error, Silly Mistake, Prompt-biased code, Missing Corner Case, Wrong Input Type, Hallucinated Object, Wrong Attribute, Incomplete Generation, and Non-Prompted Consideration. The bug patterns are presented in the form of a taxonomy. The identified bug patterns are validated using an online survey with 34 LLM practitioners and researchers. The surveyed participants generally asserted the significance and prevalence of the bug patterns. Researchers and practitioners can leverage these findings to develop effective quality assurance techniques for LLM-generated code. This study sheds light on the distinctive characteristics of LLM-generated code.

Sentiments about the reproducibility of cited papers in downstream literature offer community perspectives and have shown as a promising signal of the actual reproducibility of published findings. To train effective models to effectively predict reproducibility-oriented sentiments and further systematically study their correlation with reproducibility, we introduce the CC30k dataset, comprising a total of 30,734 citation contexts in machine learning papers. Each citation context is labeled with one of three reproducibility-oriented sentiment labels: Positive, Negative, or Neutral, reflecting the cited paper's perceived reproducibility or replicability. Of these, 25,829 are labeled through crowdsourcing, supplemented with negatives generated through a controlled pipeline to counter the scarcity of negative labels. Unlike traditional sentiment analysis datasets, CC30k focuses on reproducibility-oriented sentiments, addressing a research gap in resources for computational reproducibility studies. The dataset was created through a pipeline that includes robust data cleansing, careful crowd selection, and thorough validation. The resulting dataset achieves a labeling accuracy of 94%. We then demonstrated that the performance of three large language models significantly improves on the reproducibility-oriented sentiment classification after fine-tuning using our dataset. The dataset lays the foundation for large-scale assessments of the reproducibility of machine learning papers. The CC30k dataset and the Jupyter notebooks used to produce and analyze the dataset are publicly available at https://github.com/lamps-lab/CC30k .

Tool-using LLM agents are mostly evaluated assuming all tools work. When a tool times out, returns a week-stale value, or has its description poisoned in deployment, the developer needs a controlled way to reproduce the failure, test a fix, and confirm the fix worked before deployment. We present AgentCheck, an open-source web workbench that turns an MCP server into an intervention surface. AgentCheck runs an agent against its real tools and records every tool response, then re-runs the agent with the response perturbed by a fault (12 types) injector. Matching tool calls are replayed from cache, and later tool calls go live after the agent diverges. This yields a reproduce-intervene-confirm loop: the developer toggles a mitigation, re-runs against the identical fault, and sees if the failure goes away. Scoring has two parts: deterministic pass/fail rules, plus an LLM judge for interpretive labels, validated against human annotations. Across five agents, the best passes 105/120 scenarios and the weakest only 77. The failures are usually silent, confident use of incorrect tool outputs rather than crashes. On the weakest agent, a retry mitigation raises success on timeout error faults from as few as 30% of cases to 100%, whereas stale-data faults remain near 3-4 of 10 regardless of the mitigation. AgentCheck makes these failure modes reproducible, comparable, and verifiable before deployment.

Tool-using automation systems, from scripts and CI bots to agentic assistants, fail in recurring patterns. Common failures include unsafe side effects, invalid arguments, uncontrolled retries, and leakage of sensitive outputs. Many mitigations are model-centric and prompt-dependent, so they are brittle and do not generalize to non-LLM callers. We present Policy-First Tooling, a model-agnostic permission layer that mediates tool invocation through explicit constraints, risk-aware gating, recovery controls, and auditable explanations. The paper contributes a compact policy DSL, a runtime enforcement architecture with actionable rationale and fix hints, and a reproducible benchmark based on trace replay with controlled fault and misuse injection. In 225 controlled runs across five policy packs and three fault profiles, stricter packs improve violation prevention from 0.000 in P0 to 0.681 in P4, while task success drops from 0.356 to 0.067. Retry amplification decreases from 3.774 in P0 to 1.378 in P4, and leakage recall reaches 0.875 under injected secret outputs. These results make safety to utility trade-offs explicit and measurable.

AI agents powered by large language models exhibit strong reasoning and problem-solving capabilities, enabling them to assist scientific research tasks such as formula derivation and code generation. However, whether these agents can reliably perform end-to-end reproduction from real scientific papers remains an open question. We introduce PRBench, a benchmark of 30 expert-curated tasks spanning 11 subfields of physics. Each task requires an agent to comprehend the methodology of a published paper, implement the corresponding algorithms from scratch, and produce quantitative results matching the original publication. Agents are provided only with the task instruction and paper content, and operate in a sandboxed execution environment. All tasks are contributed by domain experts from over 20 research groups at the School of Physics, Peking University, each grounded in a real published paper and validated through end-to-end reproduction with verified ground-truth results and detailed scoring rubrics. Using an agentified assessment pipeline, we evaluate a set of coding agents on PRBench and analyze their capabilities across key dimensions of scientific reasoning and execution. The best-performing agent, OpenAI Codex powered by GPT-5.3-Codex, achieves a mean overall score of 34%. All agents exhibit a zero end-to-end callback success rate, with particularly poor performance in data accuracy and code correctness. We further identify systematic failure modes, including errors in formula implementation, inability to debug numerical simulations, and fabrication of output data. Overall, PRBench provides a rigorous benchmark for evaluating progress toward autonomous scientific research.

The way in which data are shared can affect their utility and reusability. Here, we demonstrate how data that we had previously shared in bulk can be mobilized further through a knowledge graph that allows for much more granular exploration and interrogation. The original dataset is about the computational reproducibility of GitHub-hosted Jupyter notebooks associated with biomedical publications. It contains rich metadata about the publications, associated GitHub repositories and Jupyter notebooks, and the notebooks' reproducibility. We took this dataset, converted it into semantic triples and loaded these into a triple store to create a knowledge graph, FAIR Jupyter, that we made accessible via a web service. This enables granular data exploration and analysis through queries that can be tailored to specific use cases. Such queries may provide details about any of the variables from the original dataset, highlight relationships between them or combine some of the graph's content with materials from corresponding external resources. We provide a collection of example queries addressing a range of use cases in research and education. We also outline how sets of such queries can be used to profile specific content types, either individually or by class. We conclude by discussing how such a semantically enhanced sharing of complex datasets can both enhance their FAIRness, i.e., their findability, accessibility, interoperability, and reusability, and help identify and communicate best practices, particularly with regards to data quality, standardization, automation and reproducibility.

Open-source scientific software is a major driver of scientific progress, yet its development and reuse remain difficult in collaborative settings. Researchers repeatedly face four recurring challenges: discovering and reproducing existing routines, adapting them for new use cases, sharing and scaling them across collaborators, and stabilizing them with reproducible execution environments. We present Album, an open-source framework for packaging and sharing scientific routines as executable artifacts through two minimal primitives: (i) the solution, a Python-native executable entry point that combines machine-readable metadata, arguments, environment specifications, and lifecycle hooks; and (ii) the catalog, a decentralized, git-native distribution mechanism with indexed search and optional web rendering for discovery, provenance, and governance. Album uses a two-context execution model in which a host controller evaluates manifests and prepares per-solution environments, while lifecycle hooks execute inside isolated solution environments. This design supports reproducible execution, post-environment setup, and the composition of routines with incompatible dependencies. Album can be used in conjunction with LLM agents: solutions can be drafted and revised with LLM assistance, and a MCP interface exposes cataloged solutions as callable tools for tool-grounded discovery and orchestration. We evaluate Album through four realworld imaging deployments spanning interactive visualization of electron microscopy data, integration of multiple segmentation methods, the orchestration of cryo-electron tomography competition workflows, and mineral quantification pipelines. Overall, Album complements package managers, workflow systems, and container runtimes by making scientific routines executable, shareable artifacts. Documentation and examples are available at https://album.solutions.

Scientific machine learning papers typically make computational claims, e.g., that the relative mean square error is less than 5% or that the 95% predictive credible interval covers the test data. A coding agent can be prompted to replicate those claims from paper materials alone, but the prompt does not by itself reliably preserve progress or check whether generated evidence supports the paper's claims. We introduce Paper-replication, a workflow that makes each selected paper claim a target with recorded evidence, and implement it as a coding-agent skill. The workflow makes the agent record those targets, reconstruct the paper's method, run computational experiments, link generated outputs to provenance and comparisons with the paper's claims, record where matched evidence appears in the replication report, and pass validation checks before completion. We evaluate Paper-replication on twelve independent runs across four scientific machine learning papers. All twelve workspaces pass the completion gate, and all 158 recorded targets are matched with report coverage. Even in this completed workspace state, repeated runs differ in how papers are divided into targets, in numerical fidelity to the source papers, in elapsed replication time, in the number of intermediate executions replaced before final evidence is accepted, and in the rules used to accept evidence. Paper-replication makes completion depend on workspace evidence and validation checks rather than on the agent's final message.

Quantifying bias in retrieval functions through document retrievability scores is vital for assessing recall-oriented retrieval systems. However, many studies investigating retrieval model bias lack validation of their query generation methods as accurate representations of retrievability for real users and their queries. This limitation results from the absence of established criteria for query generation in retrievability assessments. Typically, researchers resort to using frequent collocations from document corpora when no query log is available. In this study, we address the issue of reproducibility and seek to validate query generation methods by comparing retrievability scores generated from artificially generated queries to those derived from query logs. Our findings demonstrate a minimal or negligible correlation between retrievability scores from artificial queries and those from query logs. This suggests that artificially generated queries may not accurately reflect retrievability scores as derived from query logs. We further explore alternative query generation techniques, uncovering a variation that exhibits the highest correlation. This alternative approach holds promise for improving reproducibility when query logs are unavailable.

Concerns about the reproducibility of deep learning research are more prominent than ever, with no clear solution in sight. The relevance of machine learning research can only be improved if we also employ empirical rigor that incorporates reproducibility guidelines, especially so in the medical imaging field. The Medical Imaging with Deep Learning (MIDL) conference has made advancements in this direction by advocating open access, and recently also recommending authors to make their code public - both aspects being adopted by the majority of the conference submissions. This helps the reproducibility of the methods, however, there is currently little or no support for further evaluation of these supplementary material, making them vulnerable to poor quality, which affects the impact of the entire submission. We have evaluated all accepted full paper submissions to MIDL between 2018 and 2022 using established, but slightly adjusted guidelines on reproducibility and the quality of the public repositories. The evaluations show that publishing repositories and using public datasets are becoming more popular, which helps traceability, but the quality of the repositories has not improved over the years, leaving room for improvement in every aspect of designing repositories. Merely 22% of all submissions contain a repository that were deemed repeatable using our evaluations. From the commonly encountered issues during the evaluations, we propose a set of guidelines for machine learning-related research for medical imaging applications, adjusted specifically for future submissions to MIDL.

With the growing reliance on automated code completion tools in software development, the need for robust evaluation benchmarks has become critical. However, existing benchmarks focus more on code generation tasks in function and class level and provide rich text description to prompt the model. By contrast, such descriptive prompt is commonly unavailable in real development and code completion can occur in wider range of situations such as in the middle of a function or a code block. These limitations makes the evaluation poorly align with the practical scenarios of code completion tools. In this paper, we propose RepoMasterEval, a novel benchmark for evaluating code completion models constructed from real-world Python and TypeScript repositories. Each benchmark datum is generated by masking a code snippet (ground truth) from one source code file with existing test suites. To improve test accuracy of model generated code, we employ mutation testing to measure the effectiveness of the test cases and we manually crafted new test cases for those test suites with low mutation score. Our empirical evaluation on 6 state-of-the-art models shows that test argumentation is critical in improving the accuracy of the benchmark and RepoMasterEval is able to report difference in model performance in real-world scenarios. The deployment of RepoMasterEval in a collaborated company for one month also revealed that the benchmark is useful to give accurate feedback during model training and the score is in high correlation with the model's performance in practice. Based on our findings, we call for the software engineering community to build more LLM benchmarks tailored for code generation tools taking the practical and complex development environment into consideration.

Perception of toxicity evolves over time and often differs between geographies and cultural backgrounds. Similarly, black-box commercially available APIs for detecting toxicity, such as the Perspective API, are not static, but frequently retrained to address any unattended weaknesses and biases. We evaluate the implications of these changes on the reproducibility of findings that compare the relative merits of models and methods that aim to curb toxicity. Our findings suggest that research that relied on inherited automatic toxicity scores to compare models and techniques may have resulted in inaccurate findings. Rescoring all models from HELM, a widely respected living benchmark, for toxicity with the recent version of the API led to a different ranking of widely used foundation models. We suggest caution in applying apples-to-apples comparisons between studies and lay recommendations for a more structured approach to evaluating toxicity over time. Code and data are available at https://github.com/for-ai/black-box-api-challenges.

Numerous efforts have been invested in improving the effectiveness of bug localization techniques, whereas little attention is paid to making these tools run more efficiently in continuously evolving software repositories. This paper first analyzes the information retrieval model behind a classic bug localization tool, BugLocator, and builds a mathematical foundation illustrating that the model can be updated incrementally when codebase or bug reports evolve. Then, we present IncBL, a tool for Incremental Bug Localization in evolving software repositories. IncBL is evaluated on the Bugzbook dataset, and the results show that IncBL can significantly reduce the running time by 77.79% on average compared with the re-computing the model, while maintaining the same level of accuracy. We also implement IncBL as a Github App that can be easily integrated into open-source projects on GitHub. Users can deploy and use IncBL locally as well. The demo video for IncBL can be viewed at https://youtu.be/G4gMuvlJSb0, and the source code can be found at https://github.com/soarsmu/IncBL.

This study evaluates large language models (LLMs) in generating code from algorithm descriptions from recent NLP papers. The task requires two key competencies: (1) algorithm comprehension: synthesizing information from papers and academic literature to understand implementation logic, and (2) coding expertise: identifying dependencies and correctly implementing necessary APIs. To facilitate rigorous evaluation, we introduce SciReplicate-Bench, a benchmark of 100 tasks from 36 NLP papers published in 2024, featuring detailed annotations and comprehensive test cases. Building on SciReplicate-Bench, we propose Sci-Reproducer, a multi-agent framework consisting of a Paper Agent that interprets algorithmic concepts from literature and a Code Agent that retrieves dependencies from repositories and implement solutions. To assess algorithm understanding, we introduce reasoning graph accuracy, which quantifies similarity between generated and reference reasoning graphs derived from code comments and structure. For evaluating implementation quality, we employ execution accuracy, CodeBLEU, and repository dependency/API recall metrics. In our experiments, we evaluate various powerful Non-Reasoning LLMs and Reasoning LLMs as foundational models. The best-performing LLM using Sci-Reproducer achieves only 39% execution accuracy, highlighting the benchmark's difficulty.Our analysis identifies missing or inconsistent algorithm descriptions as key barriers to successful reproduction. We will open-source our benchmark, and code at https://github.com/xyzCS/SciReplicate-Bench.

A critical part of creating code suggestion systems is the pre-training of Large Language Models on vast amounts of source code and natural language text, often of questionable origin or quality. This may contribute to the presence of bugs and vulnerabilities in code generated by LLMs. While efforts to identify bugs at or after code generation exist, it is preferable to pre-train or fine-tune LLMs on curated, high-quality, and compliant datasets. The need for vast amounts of training data necessitates that such curation be automated, minimizing human intervention. We propose an automated source code autocuration technique that leverages the complete version history of open-source software projects to improve the quality of training data. This approach leverages the version history of all OSS projects to identify training data samples that have been modified or have undergone changes in at least one OSS project, and pinpoint a subset of samples that include fixes for bugs or vulnerabilities. We evaluate this method using The Stack v2 dataset, and find that 17% of the code versions in the dataset have newer versions, with 17% of those representing bug fixes, including 2.36% addressing known CVEs. The deduplicated version of Stack v2 still includes blobs vulnerable to 6,947 known CVEs. Furthermore, 58% of the blobs in the dataset were never modified after creation, suggesting they likely represent software with minimal or no use. Misidentified blob origins present an additional challenge, as they lead to the inclusion of non-permissively licensed code, raising serious compliance concerns. By addressing these issues, the training of new models can avoid perpetuating buggy code patterns or license violations. We expect our results to inspire process improvements for automated data curation, with the potential to enhance the reliability of outputs generated by AI tools.

Large language models (LLMs) are transforming automated program repair (APR) through agent-based approaches that localize bugs, generate patches, and verify fixes. However, the lack of high-quality, scalable training datasets, especially those with verifiable outputs and intermediate reasoning traces-limits progress, particularly for open-source models. In this work, we present SWE-Synth, a framework for synthesizing realistic, verifiable, and process-aware bug-fix datasets at the repository level. SWE-Synth leverages LLM agents to simulate debugging workflows, producing not only bug-fix pairs but also test cases and structured repair trajectories. Compared to manually curated datasets, our method scales with minimal human effort while preserving contextual richness and correctness. Experiments show that models trained on SWE-Synth outperform those trained on real-world datasets by 2.3% on SWE-Bench Lite. Our results highlight the potential of synthetic, agent-generated data to advance the state of the art in APR and software engineering automation.

Large Language Models (LLMs) have demonstrated impressive capabilities in code generation. However, current evaluation datasets suffer from issues such as the lack of runnable test cases, deviation from the distribution of real-world code, and the ability to evaluate only the Python language. These limitations undermine the credibility of the evaluation results. To address these limitations, we introduce MRG-Bench (Multi-language Repository-level Code Generation Benchmark), a novel dataset that provides a more accurate evaluation of LLMs in practical repository-level code generation tasks. MRG-Bench has three main features: (1) practical data sourced from real-world code repositories that align to the practical distribution, (2) multiple programming languages support, including Python, Java, and Go, and (3) project-level runnable test cases to assess the quality of the generated code. Based on MRG-Bench, we conducted extensive experiments including large language models, long-context models, and RAG-related methods. These evaluation results demonstrate that current repository-level code generation techniques suffer from significant performance deficiencies. To further investigate why models fail, we designed novel experiments to annotate the underlying causes of generation errors. The results explicitly show that the majority of methods suffer from "difficulty in understanding user requirements," failing to comprehend their assigned tasks accurately. Moreover, the impact of different repository-level contexts on this issue exhibits significant disparities across different programming languages, suggesting that, in practice, specialized contextual information needs to be designed for different languages.

We study auto research as a closed empirical loop driven by external measurement. Each submitted trial carries a hypothesis, an executable code edit, an evaluator-owned outcome, and feedback that shapes the next proposal. The output is not a generated paper or a single model checkpoint, but an auditable trajectory of proposals, code diffs, experiments, scores, and failure labels. We instantiate this loop with specialist agents that partition recipe surfaces and share measured lineage across trials. The central empirical finding is that lineage feedback lets agents turn evaluator outcomes, including crashes, budget overruns, size failures, and accuracy-gate misses, into later program-level recipe edits rather than one-shot suggestions. Across 1,197 headline-run trials plus 600 Parameter Golf control trials after one-time setup and launch, humans did not choose proposals, edit recipes, override scores, or repair failed trials during the search. In the three headline runs, the same submitted-trial loop reduces Parameter Golf validation bpb by 0.81%, raises NanoChat-D12 CORE by 38.7%, and reduces CIFAR-10 Airbench96 wallclock by 4.59%, with each task measured by its own external evaluator and legality checks. The trace includes a strict architecture-domain audit of 157 headline-run submissions and program rewrites such as a NanoChat attention-kernel path change. Within this scope the loop autonomously writes code, submits experiments, absorbs feedback, applies and combines known techniques inside each environment, and improves public starting recipes.

Existing benchmarks fail to capture a crucial aspect of intelligence: physical reasoning, the integrated ability to combine domain knowledge, symbolic reasoning, and understanding of real-world constraints. To address this gap, we introduce PhyX: the first large-scale benchmark designed to assess models capacity for physics-grounded reasoning in visual scenarios. PhyX includes 3K meticulously curated multimodal questions spanning 6 reasoning types across 25 sub-domains and 6 core physics domains: thermodynamics, electromagnetism, mechanics, modern physics, optics, and wave\&acoustics. In our comprehensive evaluation, even state-of-the-art models struggle significantly with physical reasoning. GPT-4o, Claude3.7-Sonnet, and GPT-o4-mini achieve only 32.5\%, 42.2\%, and 45.8\% accuracy respectively-performance gaps exceeding 29\% compared to human experts. Our analysis exposes critical limitations in current models: over-reliance on memorized disciplinary knowledge, excessive dependence on mathematical formulations, and surface-level visual pattern matching rather than genuine physical understanding. We provide in-depth analysis through fine-grained statistics, detailed case studies, and multiple evaluation paradigms to thoroughly examine physical reasoning capabilities. To ensure reproducibility, we implement a compatible evaluation protocol based on widely-used toolkits such as VLMEvalKit, enabling one-click evaluation.

Large language models are increasingly deployed as agents that solve tasks by interacting with external tool environments. Although recent tool-use benchmarks increasingly cover complex task settings, they still largely assume clean, stable, and trustworthy tool environments, leaving tool-environment unreliability insufficiently examined. We introduce ToolBench-X, a benchmark for evaluating agents under recoverable reliability hazards. ToolBench-X contains executable multi-step tasks across diverse domains and sequential, parallel, and mixed workflows, each paired with deterministic tools and a canonical final answer for automatic evaluation. Starting from clean tool environments, ToolBench-X injects five structured hazard types: Specification Drift, Invocation Error, Execution Failure, Output Drift, and Cross-source Conflict. Crucially, each injected instance remains solvable through at least one valid recovery path, such as retrying, fallback, verification, or cross-checking. Experiments reveal a substantial reliability gap: agents that perform well with reliable tools often fail under recoverable hazards. Further analysis shows that failures are driven less by tool-use volume or inference budget than by limited hazard diagnosis and ineffective recovery. Targeted recovery hints recover many failed tasks, while test-time scaling yields more limited gains. These results suggest that tool-use evaluation should move beyond function-call accuracy toward task completion under unreliable tool environments. The code and data is available at https://github.com/Foreverskyou/ToolBench-X.

Large Language Models (LLMs) have recently shown remarkable progress in code generation, yet their ability to construct complete software repositories from scratch remains poorly understood. A fundamental bottleneck is the lack of verifiable and scalable evaluation: existing benchmarks either focus on patch-based editing or rely on human or LLM-based judgments, which introduce bias and limit reproducibility. In this work, we present RepoZero, the first benchmark that enables fully automated, execution-based verification of repository-level generation from scratch. Our key idea is to reformulate generation as repository reproduction: given only API specifications, an agent must re-implement an entire repository such that its behavior matches the original implementation. This design allows for strict black-box validation via output equivalence, while naturally supporting large-scale construction by reusing existing open-source repositories. To further mitigate data leakage and shortcut solutions, we introduce cross-language constraints and a sandboxed evaluation protocol. Building on this benchmark, we propose an Agentic Code-Test Evolution (ACE) framework that performs iterative test generation and error-driven refinement, enabling effective test-time scaling for repository-level synthesis. Extensive experiments across multiple state-of-the-art LLMs and agent frameworks reveal that even the strongest LLM agents achieve only limited pass rates (30\% - 55\%), exposing a substantial gap between current capabilities and real-world software development requirements. Our results establish RepoZero as a challenging, scalable, and reliable testbed for end-to-end code generation, and highlight self-verification via test generation as a critical direction for advancing LLM-based coding agents.

Large Language Models (LLMs) have exhibited significant proficiency in code debugging, especially in automatic program repair, which may substantially reduce the time consumption of developers and enhance their efficiency. Significant advancements in debugging datasets have been made to promote the development of code debugging. However, these datasets primarily focus on assessing the LLM's function-level code repair capabilities, neglecting the more complex and realistic repository-level scenarios, which leads to an incomplete understanding of the LLM's challenges in repository-level debugging. While several repository-level datasets have been proposed, they often suffer from limitations such as limited diversity of tasks, languages, and error types. To mitigate this challenge, this paper introduces RepoDebug, a multi-task and multi-language repository-level code debugging dataset with 22 subtypes of errors that supports 8 commonly used programming languages and 3 debugging tasks. Furthermore, we conduct evaluation experiments on 10 LLMs, where Claude 3.5 Sonnect, the best-performing model, still cannot perform well in repository-level debugging.

AI scientist systems are increasingly deployed for autonomous research, yet their academic integrity has never been systematically evaluated. We introduce SCIINTEGRITY-BENCH, the first benchmark designed around a dilemmatic evaluation paradigm: each of its 33 scenarios across 11 trap categories is constructed so that honest acknowledgment of failure is the only correct response, while task completion requires misconduct. Across 231 evaluation runs spanning 7 state-of-the-art LLMs, the overall integrity problem rate reaches 34.2%, and no model achieves zero failures. Most strikingly, across missing-data scenarios, all seven models generate synthetic data rather than acknowledging infeasibility, differing only in whether they disclose the substitution. A further prompt ablation study separates two drivers: removing explicit completion pressure sharply reduces undisclosed fabrication from 20.6% to 3.2%, while the underlying synthesis rate remains unchanged, revealing an intrinsic completion bias that persists independent of prompt-level instructions. These findings point to the absence of honest refusal as a trained disposition as the primary driver of observed failures. We release SCIINTEGRITY-BENCH at https://github.com/liuxingtong/Sci-Integrity-Bench.

Coding agents can fix a failing example without preserving the domain rule that made it fail, so later generations can repeat the same plausible mistake. We present agentic synthesis against counterexample-supplemented sketches, a repository-native method for systems whose governing policy is discovered during implementation. A human starts with a partial, code-shaped sketch, and a coding agent generates the first implementation. When a concrete failure exposes missing or mistaken policy, an operator explicitly approves the corrected behavior and rule. The agent then revises the sketch and repairs or regenerates code and prompt surfaces for that one counterexample. The full archive preserves provenance; a selected regression set gates each revision before the next candidate is revealed; and periodic clean regeneration tests whether the evolved sketch, rather than prompt history or accumulated examples, carries the learned policy. We demonstrate the method with CatSynth, a synthetic browser application and captured coding-agent experiment. In one open-world run with GPT-5.4-mini, 8 of 14 frozen candidate cases became counterexamples. The rebuild controls inherited that promotion schedule, and all three paths passed the 8 accepted cases. Rebuilding from the evolved sketch passed 19 of 21 withheld cases, compared with 15 of 21 when rebuilding from the initial sketch and replaying all accepted examples. Retaining code across counterexamples required 9 Developer calls and 719 lines of cumulative artifact churn, versus 15 calls and 2,394 lines for replay-all, and passed 18 of 21 withheld cases. These results provide inspectable evidence that the evolved sketch carried reviewed policy and that retaining code reduced rework in this run; with one model and one reveal order, they do not establish general superiority or correctness beyond the encoded checks.

AI-assisted research is crossing a threshold: fully automated systems can now generate research papers for as little as $15, while long-horizon agents can execute experiments, draft manuscripts, and simulate critique with minimal human input. Yet this productivity frontier exposes a deeper integrity problem: under scientific pressure, even frontier LLMs still fabricate results, miss hidden errors, and fail to judge novelty reliably. Studying developments through April 2026, we present an end-to-end analysis of AI across the complete research lifecycle, organized into four epistemological phases: Creation (idea generation, literature review, coding & experiments, tables & figures), Writing (paper writing), Validation (peer review, rebuttal & revision), and Dissemination (posters, slides, videos, social media, project pages, and interactive agents). We identify a sharp, stage-dependent boundary between reliable assistance and unreliable autonomy: AI excels at structured, retrieval-grounded, and tool-mediated tasks, but remains fragile for genuinely novel ideas, research-level experiments, and scientific judgment. Generated ideas often degrade after implementation, research code lags far behind pattern-matching benchmarks, and end-to-end autonomous systems have not yet consistently reached major-venue acceptance standards. We further show that greater automation can obscure rather than eliminate failure modes, making human-governed collaboration the most credible deployment paradigm. Finally, we provide a structured taxonomy, benchmark suite, and tool inventory, cross-stage design principles, and a practitioner-oriented playbook, with resources maintained at our project page.

Automated paper reproduction has emerged as a promising approach to accelerate scientific research, employing multi-step workflow frameworks to systematically convert academic papers into executable code. However, existing frameworks often lack mechanisms to verify and refine the outputs at each generation step, or rely heavily on manually designed prompts for self-refinement, which limits their adaptability and scalability. To address these limitations, we propose a prompt-free collaborative agent framework that automatically enhances the quality of paper-to-code generation. Our approach employs two collaborative agents: a verification agent that examines whether the outputs at each step satisfy the requirements specified in the corresponding system prompt, and a refinement agent that revises the outputs based on the identified issues. Unlike previous methods that require human experts to craft specific refinement prompts for each step, our framework achieves automatic verification and improvement by leveraging only the original system prompts. We integrate our collaborative agents into the Paper2Code framework and conduct comprehensive experiments on PaperBench Code-Dev and Paper2CodeBench datasets. Experimental results demonstrate that our approach significantly improves the accuracy and completeness of reproduced code, achieving performance gains of approximately 15\% and 13\%, respectively, compared to the baseline without our agents. Furthermore, comparative experiments against Self-Refine validate the robustness and consistency of our prompt-free approach across different datasets.

Failure-inducing inputs play a crucial role in diagnosing and analyzing software bugs. Bug reports typically contain these inputs, which developers extract to facilitate debugging. Since bug reports are written in natural language, prior research has leveraged various Natural Language Processing (NLP) techniques for automated input extraction. With the advent of Large Language Models (LLMs), an important research question arises: how effectively can generative LLMs extract failure-inducing inputs from bug reports? In this paper, we propose LLPut, a technique to empirically evaluate the performance of three open-source generative LLMs -- LLaMA, Qwen, and Qwen-Coder -- in extracting relevant inputs from bug reports. We conduct an experimental evaluation on a dataset of 206 bug reports to assess the accuracy and effectiveness of these models. Our findings provide insights into the capabilities and limitations of generative LLMs in automated bug diagnosis.

Various benchmarks have been proposed to assess the performance of large language models (LLMs) in different coding scenarios. We refer to them as code-related benchmarks. However, there are no systematic guidelines by which such a benchmark should be developed to ensure its quality, reliability, and reproducibility. We propose How2Bench, which is comprised of a 55- 55-criteria checklist as a set of guidelines to govern the development of code-related benchmarks comprehensively. Using HOW2BENCH, we profiled 274 benchmarks released within the past decade and found concerning issues. Nearly 70% of the benchmarks did not take measures for data quality assurance; over 10% did not even open source or only partially open source. Many highly cited benchmarks have loopholes, including duplicated samples, incorrect reference codes/tests/prompts, and unremoved sensitive/confidential information. Finally, we conducted a human study involving 49 participants, which revealed significant gaps in awareness of the importance of data quality, reproducibility, and transparency.

Large Language Models (LLMs) have demonstrated exceptional coding capability. However, as another critical component of programming proficiency, the debugging capability of LLMs remains relatively unexplored. Previous evaluations of LLMs' debugging ability are significantly limited by the risk of data leakage, the scale of the dataset, and the variety of tested bugs. To overcome these deficiencies, we introduce `DebugBench', an LLM debugging benchmark consisting of 4,253 instances. It covers four major bug categories and 18 minor types in C++, Java, and Python. To construct DebugBench, we collect code snippets from the LeetCode community, implant bugs into source data with GPT-4, and assure rigorous quality checks. We evaluate two commercial and three open-source models in a zero-shot scenario. We find that (1) while closed-source models like GPT-4 exhibit inferior debugging performance compared to humans, open-source models such as Code Llama fail to attain any pass rate scores; (2) the complexity of debugging notably fluctuates depending on the bug category; (3) incorporating runtime feedback has a clear impact on debugging performance which is not always helpful. As an extension, we also compare LLM debugging and code generation, revealing a strong correlation between them for closed-source models. These findings will benefit the development of LLMs in debugging.

Autonomous research agents produce competitive solutions and professional-looking manuscripts, yet their outputs contain verifiability failures undetectable by surface-level evaluation: fabricated citations, unreproducible scores, and method descriptions that diverge from the implementation. We address this through three contributions. First, Chain-of-Evidence (CoE), a verifiability framework requiring every claim to be traceable to its evidence source. Second, ScientistOne, an end-to-end autonomous research system that maintains evidence chains by construction throughout literature review, solution discovery, and paper writing. Third, CoE Audit, a post-hoc audit whose four integrity checks -- score verification, specification violation, reference verification, and method-code alignment -- apply uniformly to all systems. Across 75 papers spanning five systems and five frontier research tasks, every baseline exhibits at least one systematic failure mode: hallucinated reference rates reach 21%, score verification passes in as few as 42% of papers, and method-code alignment ranges from 20% to 80%. ScientistOne achieves zero hallucinated references (0/337), perfect score verification (12/12), and the highest method-code alignment (14/15), while matching or exceeding human expert performance on all five tasks. ScientistOne further generalizes to six additional tasks spanning medical imaging, fine-grained recognition, 3D perception, and language modeling, achieving state-of-the-art on Parameter Golf and gold medals on MLE-Bench tasks where baselines fail entirely.

One-shot Program-of-Thought (PoT) emits a Python program that prints a primitive-action plan; a single invalid action silently invalidates the trajectory. We introduce RePoT (Recoverable PoT): a deterministic verified replay that walks the plan through the environment to its first invalid transition, then one LLM call that resumes from the verified prefix. RePoT costs at most one extra LLM call on the ~14% of problems where PoT fails. RePoT beats PoT by +3 to +11pp across four closed-model configurations on PuzzleZoo-775 and peaks at 96.9% vs 86.3% on gpt-5.4-mini-medium; against the matched-budget PoT-retry baseline, RePoT wins decisively on Gemini (+3.8pp, 95% CI [+2.2,+5.4]), is within sampling noise on GPT-medium and Claude, and loses on GPT-mini -- a capability-scaling pattern we begin to address with Adaptive RePoT, a rule-based dispatcher that routes between suffix repair and a fresh PoT retry based on verified-prefix length (preliminary). We replicate on PlanBench Blocksworld (+1.1 to +11.4pp) and on four open-weights models (+3.3 to +20.0pp on three of four). On Derail-550, our controlled recovery benchmark, every condition with access to checkpoint information clears >=30% on GPT-medium and >=70% on Gemini, vs <=3.1% for error-only feedback -- showing that checkpoint information, not the specific verified-prefix tail, is the load-bearing recovery signal.

AI systems can increasingly automate scientific workflows, but the reasoning that links prior evidence, generated ideas, experiments and final claims often remains implicit inside model inference. Here we introduce Xcientist, a research harness that externalizes research synthesis and experimental validation into inspectable, contract-governed processes. Xcientist organizes literature evidence, idea states, implementation plans, ablation records and repair traces as persistent research artifacts, so that generated mechanisms can be grounded, executed, tested and revised without losing their evidential basis. We identify claim drift as a failure mode of automated research, where runnable artifacts no longer support the mechanism originally claimed. Across training-free memory systems, graph-structured traffic forecasting and multi-scale physics-informed neural networks, Xcientist preserves traceable trajectories from problem formulation to mechanism design, validation and bounded revision. These results suggest that AI scientists should be evaluated not only by their final artifacts, but by whether their synthesis and validation processes remain attributable, inspectable and scientifically accountable.

One of the critical phases in software development is software testing. Testing helps with identifying potential bugs and reducing maintenance costs. The goal of automated test generation tools is to ease the development of tests by suggesting efficient bug-revealing tests. Recently, researchers have leveraged Large Language Models (LLMs) of code to generate unit tests. While the code coverage of generated tests was usually assessed, the literature has acknowledged that the coverage is weakly correlated with the efficiency of tests in bug detection. To improve over this limitation, in this paper, we introduce MuTAP for improving the effectiveness of test cases generated by LLMs in terms of revealing bugs by leveraging mutation testing. Our goal is achieved by augmenting prompts with surviving mutants, as those mutants highlight the limitations of test cases in detecting bugs. MuTAP is capable of generating effective test cases in the absence of natural language descriptions of the Program Under Test (PUTs). We employ different LLMs within MuTAP and evaluate their performance on different benchmarks. Our results show that our proposed method is able to detect up to 28% more faulty human-written code snippets. Among these, 17% remained undetected by both the current state-of-the-art fully automated test generation tool (i.e., Pynguin) and zero-shot/few-shot learning approaches on LLMs. Furthermore, MuTAP achieves a Mutation Score (MS) of 93.57% on synthetic buggy code, outperforming all other approaches in our evaluation. Our findings suggest that although LLMs can serve as a useful tool to generate test cases, they require specific post-processing steps to enhance the effectiveness of the generated test cases which may suffer from syntactic or functional errors and may be ineffective in detecting certain types of bugs and testing corner cases PUTs.

Coding agents are increasingly deployed in real software development, where a single version iteration requires months of coordinated work across many files. However, most existing benchmarks focus predominantly on single-issue bug fixes from Python repositories, with coarse pass/fail evaluation outcomes, and thus fail to capture long-horizon, multi-target development at real engineering scale. To address this gap, we present RoadmapBench, a benchmark of 115 long-horizon coding tasks grounded in real open-source version upgrades across 17 repositories and 5 programming languages. Each task places the agent on a source-version code snapshot and provides a multi-target roadmap instruction requiring it to implement the functionality introduced in the target version, with a median modification of 3,700 lines across 51 files. We conduct a systematic evaluation on thirteen frontier models and find that even the strongest, Claude-Opus-4.7, resolves only 39.1% of tasks, while the weakest achieves merely 5.2%, in stark contrast to existing bug-fix benchmarks, suggesting that long-horizon software development remains a largely unsolved problem.

With the advent of Large Language Models (LLMs), general-purpose agents have seen fundamental advancements. However, evaluating these agents presents unique challenges that distinguish them from static QA benchmarks. We observe that current agent benchmarks are heavily confounded by extraneous factors, including system prompts, toolset configurations, and environmental dynamics. Existing evaluations often rely on fragmented, researcher-specific frameworks where the prompt engineering for reasoning and tool usage varies significantly, making it difficult to attribute performance gains to the model itself. Additionally, the lack of standardized environmental data leads to untraceable errors and non-reproducible results. This lack of standardization introduces substantial unfairness and opacity into the field. We propose that a unified evaluation framework is essential for the rigorous advancement of agent evaluation. To this end, we introduce a proposal aimed at standardizing agent evaluation.

Correcting bugs using modern Automated Program Repair (APR) can be both time-consuming and resource-expensive. We describe a program repair approach that aims to improve the scalability of modern APR tools. The approach leverages program reduction in the form of program slicing to eliminate code irrelevant to fixing the bug, which improves the APR tool's overall performance. We investigate slicing's impact on all three phases of the repair process: fault localization, patch generation, and patch validation. Our empirical exploration finds that the proposed approach, on average, enhances the repair ability of the TBar APR tool, but we also discovered a few cases where it was less successful. Specifically, on examples from the widely used Defects4J dataset, we obtain a substantial reduction in median repair time, which falls from 80 minutes to just under 18 minutes. We conclude that program reduction can improve the performance of APR without degrading repair quality, but this improvement is not universal. A replication package is available via Zenodo at https://doi.org/10.5281/zenodo.13074333. Keywords: automated program repair, dynamic program slicing, fault localization, test-suite reduction, hybrid techniques.

Forecasts of explosive AI progress hinge on AI agents automating AI research. But evidence on whether agents can carry out open-ended AI research is thin. Current evaluations either test agents on narrow, verifiable tasks, which excludes open-ended research, or submit AI-generated papers to blind peer review, which is overstretched, stochastic, and suffers from poor review quality. We introduce a third way to measure progress towards AI R\&D automation. An agent takes on the central, open-ended research question of a high-quality unpublished paper, and the paper's original authors grade its output. We call these shadow evaluations. We ran shadow evaluations on two unpublished NeurIPS 2026 submissions, giving frontier agents six days and thousands of dollars of compute. The agents completed all of the engineering without human help, yet could not make substantial progress towards answering the research questions. As a result, both papers were unambiguously rejected by the authors. We identify five recurring failure modes: poor judgment about the bar for publishable research, uncreative responses to shortcomings in the research design, ineffective backtracking from dead ends, poor resource awareness, and instruction drift. A robustness check with a second model and scaffold reproduced these failures. We release the expert reviews, survey responses, agent repositories, and logs. Our results provide early evidence that today's agents can do the engineering of AI research, but struggle with critical parts of the research lifecycle.

Code agents increasingly help developers work with unfamiliar repositories, but every such task depends on a costly prerequisite: bootstrapping the repository into a usable development state. This process requires substantial trial-and-error exploration, yet the resulting knowledge--resolved dependencies, repair strategies--stays trapped in a single conversation, unavailable to future agents. We therefore formulate repository bootstrapping as a reusable startup knowledge problem and introduce BootstrapAgent, a multi-agent framework that distills the heuristics discovered during bootstrap exploration into a persistent, verifiable, agent-consumable .bootstrap contract. Through evidence extraction, structured planning, deterministic Docker-based verification, and trace-driven repair, BootstrapAgent generates a contract covering environment setup, diagnostic checks, minimal verification, and accumulated repair knowledge. We further propose warm repair with clean replay to accelerate iterative debugging without sacrificing cold-start reproducibility, and a delta repair with sanity check to prevent reward hacking. Experiments on three benchmarks show that BootstrapAgent achieves a 92.9% success rate, outperforming the baseline by over 10% while reducing downstream agent token usage by 25.9% and build time by 22.3%. Our code is available at https://github.com/Vossera/BootstrapAgent.

AI agents have the potential to aid users on a variety of consequential tasks, including conducting scientific research. To spur the development of useful agents, we need benchmarks that are challenging, but more crucially, directly correspond to real-world tasks of interest. This paper introduces such a benchmark, designed to measure the accuracy of AI agents in tackling a crucial yet surprisingly challenging aspect of scientific research: computational reproducibility. This task, fundamental to the scientific process, involves reproducing the results of a study using the provided code and data. We introduce CORE-Bench (Computational Reproducibility Agent Benchmark), a benchmark consisting of 270 tasks based on 90 scientific papers across three disciplines (computer science, social science, and medicine). Tasks in CORE-Bench consist of three difficulty levels and include both language-only and vision-language tasks. We provide an evaluation system to measure the accuracy of agents in a fast and parallelizable way, saving days of evaluation time for each run compared to a sequential implementation. We evaluated two baseline agents: the general-purpose AutoGPT and a task-specific agent called CORE-Agent. We tested both variants using two underlying language models: GPT-4o and GPT-4o-mini. The best agent achieved an accuracy of 21% on the hardest task, showing the vast scope for improvement in automating routine scientific tasks. Having agents that can reproduce existing work is a necessary step towards building agents that can conduct novel research and could verify and improve the performance of other research agents. We hope that CORE-Bench can improve the state of reproducibility and spur the development of future research agents.

Meta-learning researchers face two fundamental issues in their empirical work: prototyping and reproducibility. Researchers are prone to make mistakes when prototyping new algorithms and tasks because modern meta-learning methods rely on unconventional functionalities of machine learning frameworks. In turn, reproducing existing results becomes a tedious endeavour -- a situation exacerbated by the lack of standardized implementations and benchmarks. As a result, researchers spend inordinate amounts of time on implementing software rather than understanding and developing new ideas. This manuscript introduces learn2learn, a library for meta-learning research focused on solving those prototyping and reproducibility issues. learn2learn provides low-level routines common across a wide-range of meta-learning techniques (e.g. meta-descent, meta-reinforcement learning, few-shot learning), and builds standardized interfaces to algorithms and benchmarks on top of them. In releasing learn2learn under a free and open source license, we hope to foster a community around standardized software for meta-learning research.

As AI-generated reviews move from experimental tools into peer-review infrastructure, most robustness concerns have focused on explicit attacks such as hidden instructions and prompt injection. We study a harder and more policy-relevant failure mode: no hidden text, no prompt injection, and no changes to methods, experiments, figures, equations, proofs, or numerical results. The attacker modifies only presentation-level content, such as the abstract, contribution framing, related work, discussion, and narrative structure. We introduce adversarial repackaging: a closed-loop attack that uses AI-reviewer feedback to search for presentation-level revisions while keeping the scientific evidence fixed. Across three mainstream AI reviewers, adversarial repackaging achieves a 75.1% attack success rate and a mean score gain of +1.21/10. The effect is not explained by ordinary prose polishing. We also reveal that strategies that change how the reviewer interprets the paper, such as related-work repositioning and analytical discussion expansion, substantially outperform surface edits such as local polishing, table formatting, and algorithm boxes. Our analysis reveals two deeper structural failure modes. First, AI reviewers are easier to impress than to convince: highlighting strengths reliably increases perceived merit, while attempts to dissolve weaknesses frequently backfire. Second, AI reviewers can confuse the appearance of addressing a limitation with actually resolving it, allowing unchanged evidence to be reinterpreted as stronger scientific contribution. These results show that the deployment risk is not only malicious hidden instructions, but the emergence of paper presentation itself as an optimization surface. We release a contamination-free rolling benchmark and attack framework for testing whether AI reviewers remain anchored to scientific content under presentation-only edits.

As video games evolve into expansive, detailed worlds, visual quality becomes essential, yet increasingly challenging. Traditional testing methods, limited by resources, face difficulties in addressing the plethora of potential bugs. Machine learning offers scalable solutions; however, heavy reliance on large labeled datasets remains a constraint. Addressing this challenge, we propose a novel method, utilizing unlabeled gameplay and domain-specific augmentations to generate datasets & self-supervised objectives used during pre-training or multi-task settings for downstream visual bug detection. Our methodology uses weak-supervision to scale datasets for the crafted objectives and facilitates both autonomous and interactive weak-supervision, incorporating unsupervised clustering and/or an interactive approach based on text and geometric prompts. We demonstrate on first-person player clipping/collision bugs (FPPC) within the expansive Giantmap game world, that our approach is very effective, improving over a strong supervised baseline in a practical, very low-prevalence, low data regime (0.336 rightarrow 0.550 F1 score). With just 5 labeled "good" exemplars (i.e., 0 bugs), our self-supervised objective alone captures enough signal to outperform the low-labeled supervised settings. Building on large-pretrained vision models, our approach is adaptable across various visual bugs. Our results suggest applicability in curating datasets for broader image and video tasks within video games beyond visual bugs.