Search

Let $M$ be a holomorphically symplectic manifold, equipped with a Lagrangian fibration $\pi:\; M \to X$. A degenerate twistor deformation (sometimes also called "a Tate-Shafarevich twist") is a family of holomorphically symplectic structures on $M$ parametrized by $H^{1,1}(X)$. All members of this family are equipped with a holomorphic Lagrangian projection to $X$, and their fibers are isomorphic to the fibers of $\pi$. Assume that $M$ is a compact hyperkahler manifold of maximal holonomy, and the Lagrangian projection $\pi$ has no multiple fibers in codimension 1. We prove that $M$ has a degenerate twistor deformation $M'$ such that the Lagrangian projection $\pi:\; M' \to X$ admits a meromorphic section., Comment: 21 pages, v. 1.0

Intense highly charged ion beam production is essential for high-power heavy ion accelerators. A novel movable Vlasov launcher for superconducting high charge state Electron Cyclotron Resonance (ECR) ion source has been devised that can affect the microwave power effectiveness by a factor of about 4 in terms of highly charged ion beam production. This approach based on a dedicated microwave launching system instead of the traditional coupling scheme has led to new insight on microwave-plasma interaction. With this new understanding, the world record highly charged xenon ion beam currents have been enhanced by up to a factor of 2, which could directly and significantly enhance the performance of heavy ion accelerators and provide many new research opportunities in nuclear physics, atomic physics and other disciplines.

Nowadays, the fields of code and natural language processing are evolving rapidly. In particular, models become better at processing long context windows - supported context sizes have increased by orders of magnitude over the last few years. However, there is a shortage of benchmarks for code processing that go beyond a single file of context, while the most popular ones are limited to a single method. With this work, we aim to close this gap by introducing Long Code Arena, a suite of six benchmarks for code processing tasks that require project-wide context. These tasks cover different aspects of code processing: library-based code generation, CI builds repair, project-level code completion, commit message generation, bug localization, and module summarization. For each task, we provide a manually verified dataset for testing, an evaluation suite, and open-source baseline solutions based on popular LLMs to showcase the usage of the dataset and to simplify adoption by other researchers. We publish the benchmark page on HuggingFace Spaces with the leaderboard, links to HuggingFace Hub for all the datasets, and link to the GitHub repository with baselines: https://huggingface.co/spaces/JetBrains-Research/long-code-arena., Comment: 54 pages, 4 figures, 22 tables

Excessively long methods, loaded with multiple responsibilities, are challenging to understand, debug, reuse, and maintain. The solution lies in the widely recognized Extract Method refactoring. While the application of this refactoring is supported in modern IDEs, recommending which code fragments to extract has been the topic of many research tools. However, they often struggle to replicate real-world developer practices, resulting in recommendations that do not align with what a human developer would do in real life. To address this issue, we introduce EM-Assist, an IntelliJ IDEA plugin that uses LLMs to generate refactoring suggestions and subsequently validates, enhances, and ranks them. Finally, EM-Assist uses the IntelliJ IDE to apply the user-selected recommendation. In our extensive evaluation of 1,752 real-world refactorings that actually took place in open-source projects, EM-Assist's recall rate was 53.4% among its top-5 recommendations, compared to 39.4% for the previous best-in-class tool that relies solely on static analysis. Moreover, we conducted a usability survey with 18 industrial developers and 94.4% gave a positive rating., Comment: This paper is accepted to the tool demonstration track of the 32nd ACM Symposium on the Foundations of Software Engineering (FSE 2024). This is an author copy

In this technical report, we present three novel datasets of Kotlin code: KStack, KStack-clean, and KExercises. We also describe the results of fine-tuning CodeLlama and DeepSeek models on this data. Additionally, we present a version of the HumanEval benchmark rewritten by human experts into Kotlin - both the solutions and the tests. Our results demonstrate that small, high-quality datasets (KStack-clean and KExercises) can significantly improve model performance on code generation tasks, achieving up to a 16-point increase in pass rate on the HumanEval benchmark. Lastly, we discuss potential future work in the field of improving language modeling for Kotlin, including the use of static analysis tools in the learning process and the introduction of more intricate and realistic benchmarks.

We present a novel technique for online safety verification of autonomous systems, which performs reachability analysis efficiently for both bounded and unbounded horizons by employing neural barrier certificates. Our approach uses barrier certificates given by parameterized neural networks that depend on a given initial set, unsafe sets, and time horizon. Such networks are trained efficiently offline using system simulations sampled from regions of the state space. We then employ a meta-neural network to generalize the barrier certificates to state space regions that are outside the training set. These certificates are generated and validated online as sound over-approximations of the reachable states, thus either ensuring system safety or activating appropriate alternative actions in unsafe scenarios. We demonstrate our technique on case studies from linear models to nonlinear control-dependent models for online autonomous driving scenarios., Comment: IFAC Conference on Analysis and Design of Hybrid Systems

Modern-day Integrated Development Environments (IDEs) have come a long way from the early text editing utilities to the complex programs encompassing thousands of functions to help developers. However, with the increasing number of efficiency-enhancing tools incorporated, IDEs gradually became sophisticated software with a steep learning curve. The rise of the Large Language Models (LLMs) capable of both natural language dialogue and code generation leads to a discourse on the obsolescence of the concept of IDE. In this work, we offer a view on the place of the LLMs in the IDEs as the universal interface wrapping the IDE facilities. We envision a model that is able to perform complex actions involving multiple IDE features upon user command, stripping the user experience of the tedious work involved in searching through options and actions. For the practical part of the work, we engage with the works exploring the ability of LLMs to call for external tools to expedite a given task execution. We showcase a proof-of-concept of such a tool., Comment: First IDE Workshop, ICSE'24

Long methods that encapsulate multiple responsibilities within a single method are challenging to maintain. Choosing which statements to extract into new methods has been the target of many research tools. Despite steady improvements, these tools often fail to generate refactorings that align with developers' preferences and acceptance criteria. Given that Large Language Models (LLMs) have been trained on large code corpora, if we harness their familiarity with the way developers form functions, we could suggest refactorings that developers are likely to accept. In this paper, we advance the science and practice of refactoring by synergistically combining the insights of LLMs with the power of IDEs to perform Extract Method (EM). Our formative study on 1752 EM scenarios revealed that LLMs are very effective for giving expert suggestions, yet they are unreliable: up to 76.3% of the suggestions are hallucinations. We designed a novel approach that removes hallucinations from the candidates suggested by LLMs, then further enhances and ranks suggestions based on static analysis techniques from program slicing, and finally leverages the IDE to execute refactorings correctly. We implemented this approach in an IntelliJ IDEA plugin called EM-Assist. We empirically evaluated EM-Assist on a diverse corpus that replicates 1752 actual refactorings from open-source projects. We found that EM-Assist outperforms previous state of the art tools: EM-Assist suggests the developerperformed refactoring in 53.4% of cases, improving over the recall rate of 39.4% for previous best-in-class tools. Furthermore, we conducted firehouse surveys with 16 industrial developers and suggested refactorings on their recent commits. 81.3% of them agreed with the recommendations provided by EM-Assist.

Current state-of-the-art large language models are effective in generating high-quality text and encapsulating a broad spectrum of world knowledge. These models, however, often hallucinate and lack locally relevant factual data. Retrieval-augmented approaches were introduced to overcome these problems and provide more accurate responses. Typically, the retrieved information is simply appended to the main request, restricting the context window size of the model. We propose a novel approach for the Dynamic Retrieval-Augmented Generation (DRAG), based on the entity-augmented generation, which injects compressed embeddings of the retrieved entities into the generative model. The proposed pipeline was developed for code-generation tasks, yet can be transferred to some domains of natural language processing. To train the model, we collect and publish a new project-level code generation dataset. We use it for the evaluation along with publicly available datasets. Our approach achieves several targets: (1) lifting the length limitations of the context window, saving on the prompt size; (2) allowing huge expansion of the number of retrieval entities available for the context; (3) alleviating the problem of misspelling or failing to find relevant entity names. This allows the model to beat all baselines (except GPT-3.5) with a strong margin., Comment: 10 pages

This paper presents an optimisation-based approach for an obstacle avoidance problem within an autonomous vehicle racing context. Our control regime leverages online reachability analysis and sensor data to compute the maximal safe traversable region that an agent can traverse within the environment. The idea is to first compute a non-convex safe region, which then can be convexified via a novel coupled separating hyperplane algorithm. This derived safe area is then used to formulate a nonlinear model-predictive control problem that seeks to find an optimal and safe driving trajectory. We evaluate the proposed approach through a series of diverse experiments and assess the runtime requirements of our proposed approach through an analysis of the effects of a set of varying optimisation objectives for generating these coupled hyperplanes., Comment: In Proceedings FMAS 2023, arXiv:2311.08987

Commit messages are crucial to software development, allowing developers to track changes and collaborate effectively. Despite their utility, most commit messages lack important information since writing high-quality commit messages is tedious and time-consuming. The active research on commit message generation (CMG) has not yet led to wide adoption in practice. We argue that if we could shift the focus from commit message generation to commit message completion and use previous commit history as additional context, we could significantly improve the quality and the personal nature of the resulting commit messages. In this paper, we propose and evaluate both of these novel ideas. Since the existing datasets lack historical data, we collect and share a novel dataset called CommitChronicle, containing 10.7M commits across 20 programming languages. We use this dataset to evaluate the completion setting and the usefulness of the historical context for state-of-the-art CMG models and GPT-3.5-turbo. Our results show that in some contexts, commit message completion shows better results than generation, and that while in general GPT-3.5-turbo performs worse, it shows potential for long and detailed messages. As for the history, the results show that historical information improves the performance of CMG models in the generation task, and the performance of GPT-3.5-turbo in both generation and completion., Comment: Accepted to ASE'23. 13 pages, 5 figures

This article presents the early results of synchronous multiwavelength observations of one of the brightest gamma-ray bursts (GRBs) GRB 160625B with the detailed continuous fast optical photometry of its optical counterpart obtained by MASTER and with hard X-ray and gamma-ray emission, obtained by the Lomonosov and Konus-Wind spacecraft. The detailed photometry led us to detect the quasi-periodical emission components in the intrinsic optical emission. As a result of our analysis of synchronous multiwavelength observations, we propose a three-stage collapse scenario for this long and bright GRB. We suggest that quasiperiodic fluctuations may be associated with forced precession of a self-gravitating rapidly rotating superdense body (spinar), whose evolution is determined by a powerful magnetic field. The spinar's mass allows it to collapse into a black hole at the end of evolution., Comment: 22 pages, 12 figures, 4 tables, accepted for publication in ApJ

Competitive programming remains a very popular activity that combines both software engineering and education. In order to prepare and to practice, contestants use extensive archives of problems from past contents available on various competitive programming platforms. One way to make this process more effective is to provide an automatic tag system for the tasks. Prior works do that by either using the tasks' problem statements or the code of their solutions. In this study, we investigate which information source is more valuable for tag prediction. To answer that question, we compare existing approaches of both types on the same dataset and with the same set of tags. Then, we propose a novel approach, which is an ensemble of the Gated Graph Neural Network model for analyzing solutions and the Bidirectional Encoder Representations from Transformers model for processing statements. Our experiments show that our approach outperforms previously proposed models by 0.175 of the PR-AUC metric., Comment: The work was carried out at the end of 2020. 7 pages, 2 figures

There has been increasing interest in studying the effect of giving births to unintended pregnancies on later life physical and mental health. In this article, we provide the protocol for our planned observational study on the long-term mental and physical health consequences for mothers who bear children resulting from unintended pregnancies. We aim to use the data from the Wisconsin Longitudinal Study (WLS) and examine the effect of births from unintended pregnancies on a broad range of outcomes, including mental depression, psychological well-being, physical health, alcohol usage, and economic well-being. To strengthen our causal findings, we plan to address our research questions on two subgroups, Catholics and non-Catholics, and discover the "replicable" outcomes for which the effect of unintended pregnancy is negative (or, positive) in both subgroups. Following the idea of non-random cross-screening, the data will be split according to whether the woman is Catholic or not, and then one part of the data will be used to select the hypotheses and design the corresponding tests for the second part of the data. In past use of cross-screening (automatic cross-screening) there was only one team of investigators that dealt with both parts of the data so that the investigators would need to decide on an analysis plan before looking at the data. In this protocol, we describe plans to carry out a novel flexible cross-screening in which there will be two teams of investigators with access only to one part of data and each team will use their part of the data to decide how to plan the analysis for the second team's data. In addition to the above replicability analysis, we also discuss the plan to test the global null hypothesis that is intended to identify the outcomes which are affected by unintended pregnancy for at least one of the two subgroups of Catholics and non-Catholics., Comment: 29 pages

Meta-analysis is routinely performed in many scientific disciplines. This analysis is attractive since discoveries are possible even when all the individual studies are underpowered. However, the meta-analytic discoveries may be entirely driven by signal in a single study, and thus non-replicable. Although the great majority of meta-analyses carried out to date do not infer on the replicability of their findings, it is possible to do so. We provide a selective overview of analyses that can be carried out towards establishing replicability of the scientific findings. We describe methods for the setting where a single outcome is examined in multiple studies (as is common in systematic reviews of medical interventions), as well as for the setting where multiple studies each examine multiple features (as in genomics applications). We also discuss some of the current shortcomings and future directions., Comment: 30 pages

The RELEC scientific instrumentation onboard the Vernov spacecraft launched on July 8, 2014, included the DRGE gamma-ray and electron spectrometer. This instrument incorporates a set of scintillation phoswich detectors, including four identical X-ray and gamma-ray detectors in the energy range from 10 keV to 3 MeV with a total area of $\sim$500 $cm^{2}$ directed toward the nadir, and an electron spectrometer containing three mutually orthogonal detector units with a geometry factor of $\sim$2 $cm^{2} sr$, which is also sensitive to X-rays and gamma-rays. The goal of the space experiment with the DRGE instrument was to investigate phenomena with fast temporal variability, in particular, terrestrial gamma-ray flashes (TGFs) and magnetospheric electron precipitations. However, the detectors of the DRGE instrument could record cosmic gamma-ray bursts (GRBs) and allowed one not only to perform a detailed analysis of the gamma-ray variability but also to compare the time profiles with the measurements made by other instruments of the RELEC scientific instrumentation (the detectors of optical and ultraviolet flashes, the radio-frequency and low-frequency analyzers of electromagnetic field parameters). We present the results of our observations of cosmic GRB 141011A and GRB 141104A, compare the parameters obtained in the GBM/Fermi and KONUSWind experiments, and estimate the redshifts and Eiso for the sources of these GRBs. The detectability of GRBs and good agreement between the independent estimates of their parameters obtained in various experiments are an important factor of the successful operation of similar detectors onboard the Lomonosov spacecraft., Comment: 19 pages, 8 figures

In software engineering, different approaches and machine learning models leverage different types of data: source code, textual information, historical data. An important part of any project is its dependencies. The list of dependencies is relatively small but carries a lot of semantics with it, which can be used to compare projects or make judgements about them. In this paper, we focus on Python projects and their PyPi dependencies in the form of requirements.txt files. We compile a dataset of 7,132 Python projects and their dependencies, as well as use Git to pull their versions from previous years. Using this data, we build 32-dimensional embeddings of libraries by applying Singular Value Decomposition to the co-occurrence matrix of projects and libraries. We then cluster the embeddings and study their semantic relations. To showcase the usefulness of such lightweight library embeddings, we introduce a prototype tool for suggesting relevant libraries to a given project. The tool computes project embeddings and uses dependencies of projects with similar embeddings to form suggestions. To compare different library recommenders, we have created a benchmark based on the evolution of dependency sets in open-source projects. Approaches based on the created embeddings significantly outperform the baseline of showing the most popular libraries in a given year. We have also conducted a user study that showed that the suggestions differ in quality for different project domains and that even relevant suggestions might be not particularly useful. Finally, to facilitate potentially more useful recommendations, we extended the recommender system with an option to suggest rarer libraries., Comment: The work was carried out at the end of 2020. 11 pages, 4 figures

In recent years, researchers have created and introduced a significant number of various code generation models. As human evaluation of every new model version is unfeasible, the community adopted automatic evaluation metrics such as BLEU to approximate the results of human judgement. These metrics originate from the machine translation domain and it is unclear whether they are applicable for the code generation tasks and how well they agree with the human evaluation on this task. There are also other metrics, CodeBLEU and RUBY, developed to estimate the similarity of code, that take into account the properties of source code. However, for these metrics there are hardly any studies on their agreement with the human evaluation. Despite all that, minimal differences in the metric scores have been used in recent papers to claim superiority of some code generation models over the others. In this paper, we present a study on the applicability of six metrics -- BLEU, ROUGE-L, METEOR, ChrF, CodeBLEU, and RUBY -- for evaluation of code generation models. We conduct a study on two different code generation datasets and use human annotators to assess the quality of all models run on these datasets. The results indicate that for the CoNaLa dataset of Python one-liners, none of the metrics can correctly emulate human judgement on which model is better with >95% certainty if the difference in model scores is less than 5 points. For the HearthStone dataset, which consists of classes of a particular structure, a difference in model scores of at least 2 points is enough to claim the superiority of one model over the other. Our findings suggest that the ChrF metric is a better fit for the evaluation of code generation models than the commonly used BLEU and CodeBLEU. Yet, finding a metric for code generation that closely agrees with humans requires additional work.