We introduce DSCodeBench, a new benchmark designed to evaluate large language models (LLMs) on complicated and realistic data science code generation tasks. DSCodeBench consists of 1,000 carefully constructed problems sourced from realistic problems from GitHub across ten widely used Python data science libraries. DSCodeBench offers a more challenging and representative testbed, more complex code solutions, more comprehensive data science libraries, clearer and better structured problem descriptions, and stronger test suites. To construct the DSCodeBench, we develop a robust pipeline that combines task scope selection, code construction, test case generation, and problem description synthesis. The process is paired with rigorous manual editing to ensure alignment and enhance the reliability of the evaluation. Experimental result shows that DSCodeBench exhibits robust scaling behavior, where larger models systematically outperform smaller ones, validating its ability to distinguish model capabilities. The best LLM we test, GPT-4o, has a pass@1 of 0.392, indicating that LLMs still have a large room to improve for realistic data science code generation tasks. We believe DSCodeBench will serve as a rigorous and trustworthy foundation for advancing LLM-based data science programming.

The prevailing single-source paradigm in wireless sensing produces specialized models that are unscalable and generalize poorly to new tasks. Multi-source pre-training offers a path toward a generalist backbone but poses challenges including task heterogeneity, data redundancy, structural incompatibility, and the lack of a general-purpose pre-training objective. To address these issues, we propose WiSwiss, a comprehensive self-supervised multi-source pre-training framework that learns a general-purpose backbone for each modality. WiSwiss integrates semantic deduplication for dataset curation and a transformation-invariant pre-training objective. Experiments show that WiSwiss outperforms models trained from scratch, improving WiFi and mmWave performance by 4.5% and 10.3%, respectively, while reducing fine-tuning data requirements by 22.2% and 28.6%. We also present a qualitative study of scaling laws, showing that gains are task-dependent and that larger models require sufficiently large and diverse pre-training corpora to achieve substantial improvements.

Lean is an advanced proof assistant designed to facilitate formal theorem proving by providing a variety of interactive feedback. In this paper, we explore methodologies to leverage proof assistant feedback to augment the capabilities of large language models in constructing formal proofs. First, we deploy online reinforcement learning using Lean verification outcomes as the reward signal to improve the proof completion policy. This straightforward approach shows great promise in enhancing the model's alignment with the formal verification system. In addition, we propose RMaxTS, a variant of Monte-Carlo tree search that employs an intrinsic-reward-driven exploration strategy to generate diverse proof paths. The tree structure is organized to represent the transitions of intermediate tactic states, extracted from the compilation messages given by Lean's tactic mode. The intrinsic reward is constructed to incentivize the discovery of novel tactic states, which helps to to mitigate the sparse-reward problem inherent in proof search. These techniques lead to a more efficient planning scheme for formal proof generation, achieving new state-of-the-art results on both miniF2F and ProofNet benchmarks.

In millimeter-wave and terahertz band communication systems, precise beam prediction is crucial for optimizing network performance and enhancing signal transmission efficiency. Traditional beam prediction methods have primarily relied on single-modal data, which often fails to capture the comprehensive environmental information necessary for optimal accuracy. In contrast, multi-modal data-based approaches offer a more promising solution by leveraging the strengths of diverse data sources. However, many existing fusion methods are static, inadequately accounting for variations in information content across different modalities, which can hinder the full utilization of each modality’s advantages. To address these limitations, this paper proposes an advanced multi-modal beam prediction method that integrates multipath-like data augmentation (MLDA), cross-modal feature enhancement (CMFE), and an uncertainty-aware dynamic fusion mechanism. Our approach combines image and radar data to predict beam indices, dynamically adjusting the weights of different modalities to accommodate varying information densities. The proposed method employs ResNet34 for feature extraction from the multi-modal data, followed by a cross-modal feature enhancement module that aggregates complementary information from the image and radar data. Finally, the dynamic fusion mechanism integrates the predictions from the single-modal data. Experimental results demonstrate that our method significantly improves the accuracy and robustness of beam prediction, achieving an overall accuracy of 89.72%. The performance of the proposed method is further validated through comparisons with various existing methods and comprehensive ablation studies, highlighting its superiority in multi-modal assisted beam prediction scenarios.

Mutation faults are the core of mutation testing and have been widely used in many software testing tasks. Hence, efficiently constructing high-quality mutation faults is critical. To address the effectiveness limitations of traditional and deep learning-based mutation techniques, we first proposed LEAM , utilizing a syntax-guided encoder–decoder architecture with extended grammar rules. While LEAM significantly enhances the effectiveness, it does not consider the associated testing cost. To further improve the efficiency of LEAM , we propose LEAM++ , adopting a novel selective mutation fault construction module based on the probability of grammar rule sequences and the similarity of mutation faults. We extensively evaluate LEAM++ using Defects4J. Regarding effectiveness, the results demonstrate that the mutation faults constructed by LEAM++ can better represent real faults than two traditional techniques ( Major and PIT ) and the deep learning-based technique ( DeepMutation ), and substantially boost three downstream applications, i.e., mutation-based test case prioritization, mutation-based fault localization, and mutation-based bug detection. Regarding efficiency, LEAM++ demonstrates superiority over the four selective mutation testing techniques across three scenarios, i.e., mutation testing, mutation-based test case prioritization, and mutation-based fault localization. Our work serves as an important step toward the efficiently automated construction of mutation faults.

Code pre-trained models have shown promising effectiveness in various software engineering tasks. Among these tasks, many tasks are related to software evolution and/or code editing. However, existing code pre-trained models often overlook the real-world code editing data and the evolutionary nature of the editing process. In this paper, to simulate the step-by-step code editing process of human developers, we propose DivoT5, a pre-trained model based on directional diffusion at the data level. In DivoT5, we adopt two categories of pre-training tasks. The first category is mask and denoising tasks augmented with a diffusion direction representing code evolution. That is, we first apply a noising process to the code snippets before evolution, and then ask the pre-training process to restore the snippets with noise into the code snippets after evolution. The second category is tasks aiming to reinforce the evolutionary direction. That is, we first generate various intermediate versions for each pair of snippets before and after evolution, and then ask the pre-training process to transform the intermediate versions into the snippet after evolution for each pair. We evaluate DivoT5 for two code-editing scenarios (including a number of tasks) and one non-editing scenario using four downstream tasks. For each downstream task, we fine-tune the pre-trained DivoT5 on multiple corresponding datasets and evaluate its effectiveness across diverse scenarios Our experimental results show that ivoT5 achieves state-of-the-art (SOTA) performance on most tasks in comparison to models of the same scale (220M), large-scale (770M, 6.7B) models in fine-tuning, and billion-scale (6.7B, 8B, ChatGPT) instruct models in few-shot settings. For one code-editing task (i.e., CodeReview in NL-based CodeRefinement task), DivoT5 pre-trained on top of CodeT5-small (60M) can even outperform CodeT5-base (220M) and other pre-trained models with 220M parameters except for DivoT5 pre-trained on top of CodeT5-base (220M).

GUI test case migration is to migrate GUI test cases from a source app to a target app. The key of test case migration is widget matching. Recently, researchers have proposed various approaches by formulating widget matching as a matching task. However, since these matching approaches depend on static word embeddings without using contextual information to represent widgets and manually formulated matching functions, there are main limitations of these matching approaches when handling complex matching relations in apps. To address the limitations, we propose the first learning-based widget matching approach named TEMdroid (TEst Migration) for test case migration. Unlike the existing approaches, TEMdroid uses BERT to capture contextual information and learns a matching model to match widgets. Additionally, to balance the significant imbalance between positive and negative samples in apps, we design a two-stage training strategy where we first train a hard-negative sample miner to mine hard-negative samples, and further train a matching model using positive samples and mined hard-negative samples. Our evaluation on 34 apps shows that TEM-droid is effective in event matching (i.e., widget matching and target event synthesis) and test case migration. For event matching, TEM-droid's Top1 accuracy is 76%, improving over 17% compared to baselines. For test case migration, TEMdroid's F1 score is 89%, also 7% improvement compared to the baseline approach.

Pretrained models for code have exhibited promising performance across various code-related tasks, such as code summarization, code completion, code translation, and bug detection. However, despite their success, the majority of current models still represent code as a token sequence, which may not adequately capture the essence of the underlying code structure.

GUI test case migration is the process of migrating GUI test cases from a source app to a target app for a specific functionality. However, test cases obtained via existing migration approaches can hardly be directly used to test target functionalities and typically require additional manual modifications. This problem may significantly impact the effectiveness of testing target functionalities and the practical applicability of migration approaches. In this paper, we propose MigratePro, the first approach to enhancing GUI test case migration via synthesizing a new test case based on multiple test cases for the same functionality migrated from various source apps to the target app. The aim of MigratePro is to produce functional test cases with less human intervention. Specifically, given multiple migrated test cases for the same functionality in the target app, MigratePro first combines all the GUI states related to these migrated test cases into an overall state-sequence. Then, MigratePro organizes events and assertions from migrated test cases according to the overall state-sequence and endeavors to remove the should-be-removed events and assertions, while also incorporating some connection events in order to make the should-be-included events and assertions executable. Our evaluation on 30 apps, 34 functionalities, and 127 test cases shows that MigratePro improves the capability of three representative migration approaches (i.e., Craftdroid, AppFlow, ATM), successfully improving testing the target functionalities by 86%, 333%, and 300%, respectively. These results underscore the generalizability of MigratePro for effectively enhancing migration approaches.

Unit testing is an essential activity in software development for verifying the correctness of software components. However, manually writing unit tests is challenging and time-consuming. The emergence of Large Language Models (LLMs) offers a new direction for automating unit test generation. Existing research primarily focuses on closed-source LLMs (e.g., ChatGPT and CodeX) with fixed prompting strategies, leaving the capabilities of advanced open-source LLMs with various prompting settings unexplored. Particularly, open-source LLMs offer advantages in data privacy protection and have demonstrated superior performance in some tasks. Moreover, effective prompting is crucial for maximizing LLMs' capabilities. In this paper, we conduct the first empirical study to fill this gap, based on 17 Java projects, five widely-used open-source LLMs with different structures and parameter sizes, and comprehensive evaluation metrics. Our findings highlight the significant influence of various prompt factors, show the performance of open-source LLMs compared to the commercial GPT-4 and the traditional Evosuite, and identify limitations in LLM-based unit test generation. We then derive a series of implications from our study to guide future research and practical use of LLM-based unit test generation.

Automated program repair (APR) aims to reduce the effort of software development. With the development of deep learning, lots of DL-based APR approaches have been proposed using an encoder-decoder architecture. Despite the promising performance, these models share the same limitation: generating lots of untypable patches. The main reason for this phenomenon is that the existing models do not consider the constraints of code captured by a set of typing rules. In this paper, we propose, Tare, a type-aware model for neural program repair to learn the typing rules. To encode an individual typing rule, we introduce three novel components: (1) a novel type of grammars, T-Grammar, that integrates the type information into a standard grammar, (2) a novel representation of code, T-Graph, that integrates the key information needed for type checking an AST, and (3) a novel type-aware neural program repair approach, Tare, that encodes the T-Graph and generates the patches guided by T-Grammar. The experiment was conducted on three benchmarks, 393 bugs from Defects4J v1.2, 444 additional bugs from Defects4J v2.0, and 40 bugs from QuixBugs. Our results show that Tare repairs 62, 32, and 27 bugs on these benchmarks respectively, and outperforms the existing APR approaches on all benchmarks. Further analysis also shows that Tare tends to generate more compilable patches than the existing DL-based APR approaches with the typing rule information.

Delta debugging provides an efficient and systematic approach to isolate and identify a minimal subsequence that exhibit a specific property. A notable trend in the development of delta debugging is to address data with domain-specific structures, such as programs. However, the efficiency and effectiveness of domain-specific delta debugging algorithms still present challenges. Probabilistic delta debugging (ProbDD) enhances the ddmin algorithm, which forms the foundation of most domain-specific delta debugging approaches, by incorporating a probabilistic model. By replacing the ddmin component with ProbDD, algorithms relying on ddmin can achieve superior performance. Meanwhile, domain-specific delta debugging techniques, such as Perses, have been designed to cater to the abstract syntax tree (AST) and follow predefined sequences of attempts to minimize programs. These techniques benefit from the use of AST-based transformations, enabling them to achieve even smaller results efficiently. However, we observe that ProbDD assumes independence between elements, which may limit their performance in capturing syntactic relationships. Additionally, domain-specific approaches such as Perses rely on a predefined sequence of attempts the removal of the element and fail to utilize the information from existing test results.In this paper, we propose T-PDD, a novel approach that addresses these limitations. T-PDD leverages the AST to construct a probabilistic model, both utilizing historical test results and capturing syntactic relationships to estimate the probabilities of elements being retained in the result. It selects a set of elements that maximizes the gain for the next test based on the model and updates the model using the test results.In our evaluation, we assess our approach on 107 real-world subjects. The results demonstrate an average improvement of 26.95% in processing time and a 3.4x reduction in result size compared to Perses in the best-case scenario.

Collaborative development is critical to improve the productivity. Multiple contributors work simultaneously on the same project and might make changes to the same code locations. This can cause conflicts and require manual intervention from developers to resolve them. To alleviate the human efforts of manual conflict resolution, researchers have proposed various automatic techniques. More recently, deep learning models have been adopted to solve this problem and achieved state-of-the-art performance. However, these techniques leverage classification to combine the existing elements of input. The classification- based models cannot generate new tokens or produce flexible combinations, and have a wrong hypothesis that fine-grained conflicts of one single coarse-grained conflict are independent. In this work, we propose to generate the resolutions of merge conflicts from a totally new perspective, that is, generation, and we present a conflict resolution technique, MergeGen. First, we design a structural and fine-grained conflict-aware representation for the merge conflicts. Then, we propose to leverage an encoder- decoder-based generative model to process the designed conflict representation and generate the resolutions auto-regressively. We further perform a comprehensive study to evaluate the effectiveness of MergeGen. The quantitative results show that MergeGen outperforms the state-of-the-art (SOTA) techniques from both precision and accuracy. Our evaluation on multiple programming languages verifies the good generalization ability of MergeGen. In addition, the ablation study shows that the major component of our technique makes a positive contribution to the performance of MergeGen, and the granularity analysis reveals the high tolerance of MergeGen to coarse-grained conflicts. Moreover, the analysis on generating new tokens further proves the advance of generative models.

Existing graph neural networks (GNNs) largely rely on node embeddings, which represent a node as a vector by its identity, type, or content. However, graphs with unattributed nodes widely exist in real-world applications (e.g., anonymized social networks). Previous GNNs either assign random labels to nodes (which introduces artefacts to the GNN) or assign one embedding to all nodes (which fails to explicitly distinguish one node from another). Further, when these GNNs are applied to unattributed node classification problems, they have an undesired equivariance property, which are fundamentally unable to address the data with multiple possible outputs. In this paper, we analyze the limitation of existing approaches to node classification problems. Inspired by our analysis, we propose a generalized equivariance property and a Preferential Labeling technique that satisfies the desired property asymptotically. Experimental results show that we achieve high performance in several unattributed node classification tasks.

Commit messages summarize code changes of each commit in natural language, which help developers understand code changes without digging into detailed implementations and play an essential role in comprehending software evolution. To alleviate human efforts in writing commit messages, researchers have proposed various automated techniques to generate commit messages, including template-based, information retrieval-based, and learning-based techniques. Although promising, previous techniques have limited effectiveness due to their coarse-grained code change representations.

Recently, neural techniques have been used to generate source code automatically. While promising for declarative languages, these approaches achieve much poorer performance on datasets for imperative languages. Since a declarative language is typically embedded in an imperative language (i.e., the turducken-style programming) in real-world software development, the promising results on declarative languages can hardly lead to significant reduction of manual software development efforts. In this paper, we define a new code generation task: given a natural language comment, this task aims to generate a program in a base imperative language with an embedded declarative language. To our knowledge, this is the first turducken-style code generation task. For this task, we present Lyra: a dataset in Python with embedded SQL. This dataset contains 2,000 carefully annotated database manipulation programs from real usage projects. Each program is paired with both a Chinese comment and an English comment. In our experiment, we adopted Transformer, BERT-style, and GPT-style models as baselines. In the best setting, GPT-style model can achieve 24% and 25.5% AST exact matching accuracy using Chinese and English comments, respectively. Therefore, we believe that Lyra provides a new challenge for code generation. Yet, overcoming this challenge may significantly boost the applicability of code generation techniques for real-world software development.

Word embedding has been widely used in various areas to boost the performance of the neural models. However, when processing context-free languages, embedding grammar rules with word embedding loses two types of information. One is the structural relationship between the grammar rules, and the other one is the content information of the rule definition. In this paper, we make the first attempt to learn a grammar-preserving rule embedding. We first introduce a novel graph structure to represent the context-free grammar. Then, we apply a Graph Neural Network (GNN) to extract the structural information and use a gating layer to integrate content information. We conducted experiments on six widely-used benchmarks containing four context-free languages. The results show that our approach improves the accuracy of the base model by 0.8 to 6.4 percentage points. Furthermore, Grape also achieves 1.6 F1 score improvement on the method naming task which shows the generality of our approach.

Mutation faults are the core of mutation testing and have been widely used in many other software testing and debugging tasks. Hence, constructing high-quality mutation faults is critical. There are many traditional mutation techniques that construct syntactic mutation faults based on a limited set of manually-defined mutation operators. To improve them, the state-of-the-art deep-learning (DL) based technique (i.e., DeepMutation) has been proposed to construct mutation faults by learning from real faults via classic sequence-to-sequence neural machine translation (NMT). However, its performance is not satisfactory since it cannot ensure syntactic correctness of constructed mutation faults and suffers from the effectiveness issue due to the huge search space and limited features by simply treating each targeted method as a token stream.

Coverage-based fault localization has been extensively studied in the literature due to its effectiveness and lightweightness for real-world systems. However, existing techniques often utilize coverage in an oversimplified way by abstracting detailed coverage into numbers of tests or boolean vectors, thus limiting their effectiveness in practice. In this work, we present a novel coverage-based fault localization technique, GRACE, which fully utilizes detailed coverage information with graph-based representation learning. Our intuition is that coverage can be regarded as connective relationships between tests and program entities, which can be inherently and integrally represented by a graph structure: with tests and program entities as nodes, while with coverage and code structures as edges. Therefore, we first propose a novel graph-based representation to reserve all detailed coverage information and fine-grained code structures into one graph. Then we leverage Gated Graph Neural Network to learn valuable features from the graph-based coverage representation and rank program entities in a listwise way. Our evaluation on the widely used benchmark Defects4J (V1.2.0) shows that GRACE significantly outperforms state-of-the-art coverage-based fault localization: GRACE localizes 195 bugs within Top-1 whereas the best compared technique can at most localize 166 bugs within Top-1. We further investigate the impact of each GRACE component and find that they all positively contribute to GRACE. In addition, our results also demonstrate that GRACE has learnt essential features from coverage, which are complementary to various information used in existing learning-based fault localization. Finally, we evaluate GRACE in the cross-project prediction scenario on extra 226 bugs from Defects4J (V2.0.0), and find that GRACE consistently outperforms state-of-the-art coverage-based techniques.

Automated Program Repair (APR) helps improve the efficiency of software development and maintenance. Recent APR techniques use deep learning, particularly the encoder-decoder architecture, to generate patches. Though existing DL-based APR approaches have proposed different encoder architectures, the decoder remains to be the standard one, which generates a sequence of tokens one by one to replace the faulty statement. This decoder has multiple limitations: 1) allowing to generate syntactically incorrect programs, 2) inefficiently representing small edits, and 3) not being able to generate project-specific identifiers.

A code generation system generates programming language code based on an input natural language description. State-of-the-art approaches rely on neural networks for code generation. However, these code generators suffer from two problems. One is the long dependency problem, where a code element often depends on another far-away code element. A variable reference, for example, depends on its definition, which may appear quite a few lines before. The other problem is structure modeling, as programs contain rich structural information. In this paper, we propose a novel tree-based neural architecture, TreeGen, for code generation. TreeGen uses the attention mechanism of Transformers to alleviate the long-dependency problem, and introduces a novel AST reader (encoder) to incorporate grammar rules and AST structures into the network. We evaluated TreeGen on a Python benchmark, HearthStone, and two semantic parsing benchmarks, ATIS and GEO. TreeGen outperformed the previous state-of-the-art approach by 4.5 percentage points on HearthStone, and achieved the best accuracy among neural network-based approaches on ATIS (89.1%) and GEO (89.6%). We also conducted an ablation test to better understand each component of our model.

The Boolean satisfiability problem (SAT) is a famous NP-complete problem in computer science. An effective way for solving a satisfiable SAT problem is the stochastic local search (SLS). However, in this method, the initialization is assigned in a random manner, which impacts the effectiveness of SLS solvers. To address this problem, we propose NLocalSAT. NLocalSAT combines SLS with a solution prediction model, which boosts SLS by changing initialization assignments with a neural network. We evaluated NLocalSAT on five SLS solvers (CCAnr, Sparrow, CPSparrow, YalSAT, and probSAT) with instances in the random track of SAT Competition 2018. The experimental results show that solvers with NLocalSAT achieve 27% ~ 62% improvement over the original SLS solvers.

Code retrieval helps developers reuse code snippets in the open-source projects. Given a natural language description, code retrieval aims to search for the most relevant code relevant among a set of code snippets. Existing state-of-the-art approaches apply neural networks to code retrieval. However, these approaches still fail to capture an important feature: overlaps. The overlaps between different names used by different people indicate that two different names may be potentially related (e.g., "message" and "msg"), and the overlaps between identifiers in code and words in natural language descriptions indicate that the code snippet and the description may potentially be related.

Code generation maps a program description to executable source code in a programming language. Existing approaches mainly rely on a recurrent neural network (RNN) as the decoder. However, we find that a program contains significantly more tokens than a natural language sentence, and thus it may be inappropriate for RNN to capture such a long sequence. In this paper, we propose a grammar-based structural convolutional neural network (CNN) for code generation. Our model generates a program by predicting the grammar rules of the programming language; we design several CNN modules, including the tree-based convolution and pre-order convolution, whose information is further aggregated by dedicated attentive pooling layers. Experimental results on the HearthStone benchmark dataset show that our CNN code generator significantly outperforms the previous state-of-the-art method by 5 percentage points; additional experiments on several semantic parsing tasks demonstrate the robustness of our model. We also conduct in-depth ablation test to better understand each component of our model.