Ying Zou

Kimi Team, Tongtong Bai, Yifan Bai et al.

We introduce Kimi K2.5, an open-source multimodal agentic model designed to advance general agentic intelligence. K2.5 emphasizes the joint optimization of text and vision so that two modalities enhance each other. This includes a series of techniques such as joint text-vision pre-training, zero-vision SFT, and joint text-vision reinforcement learning. Building on this multimodal foundation, K2.5 introduces Agent Swarm, a self-directed parallel agent orchestration framework that dynamically decomposes complex tasks into heterogeneous sub-problems and executes them concurrently. Extensive evaluations show that Kimi K2.5 achieves state-of-the-art results across various domains including coding, vision, reasoning, and agentic tasks. Agent Swarm also reduces latency by up to $4.5\times$ over single-agent baselines. We release the post-trained Kimi K2.5 model checkpoint to facilitate future research and real-world applications of agentic intelligence.

Hejun Huang, Zuguo Chen, Ying Zou et al.

Characteristics such as low contrast and significant organ shape variations are often exhibited in medical images. The improvement of segmentation performance in medical imaging is limited by the generally insufficient adaptive capabilities of existing attention mechanisms. An efficient Channel Prior Convolutional Attention (CPCA) method is proposed in this paper, supporting the dynamic distribution of attention weights in both channel and spatial dimensions. Spatial relationships are effectively extracted while preserving the channel prior by employing a multi-scale depth-wise convolutional module. The ability to focus on informative channels and important regions is possessed by CPCA. A segmentation network called CPCANet for medical image segmentation is proposed based on CPCA. CPCANet is validated on two publicly available datasets. Improved segmentation performance is achieved by CPCANet while requiring fewer computational resources through comparisons with state-of-the-art algorithms. Our code is publicly available at \url{https://github.com/Cuthbert-Huang/CPCANet}.

Hejun Huang, Zuguo Chen, Chaoyang Chen et al.

A network based on complementary consistency training, called CC-Net, has been proposed for semi-supervised left atrium image segmentation. CC-Net efficiently utilizes unlabeled data from the perspective of complementary information to address the problem of limited ability of existing semi-supervised segmentation algorithms to extract information from unlabeled data. The complementary symmetric structure of CC-Net includes a main model and two auxiliary models. The complementary model inter-perturbations between the main and auxiliary models force consistency to form complementary consistency. The complementary information obtained by the two auxiliary models helps the main model to effectively focus on ambiguous areas, while enforcing consistency between the models is advantageous in obtaining decision boundaries with low uncertainty. CC-Net has been validated on two public datasets. In the case of specific proportions of labeled data, compared with current advanced algorithms, CC-Net has the best semi-supervised segmentation performance. Our code is publicly available at https://github.com/Cuthbert-Huang/CC-Net.

Suzhen Zhong, Shayan Noei, Ying Zou et al.

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

Maram Assi, Safwat Hassan, Ying Zou

The exponential growth of the mobile app market underscores the importance of constant innovation and rapid response to user demands. As user satisfaction is paramount to the success of a mobile application (app), developers typically rely on user reviews, which represent user feedback that includes ratings and comments to identify areas for improvement. However, the sheer volume of user reviews poses challenges in manual analysis, necessitating automated approaches. Existing automated approaches either analyze only the target apps reviews, neglecting the comparison of similar features to competitors or fail to provide suggestions for feature enhancement. To address these gaps, we propose a Large Language Model (LLM)-based Competitive User Review Analysis for Feature Enhancement) (LLM-Cure), an approach powered by LLMs to automatically generate suggestion s for mobile app feature improvements. More specifically, LLM-Cure identifies and categorizes features within reviews by applying LLMs. When provided with a complaint in a user review, LLM-Cure curates highly rated (4 and 5 stars) reviews in competing apps related to the complaint and proposes potential improvements tailored to the target application. We evaluate LLM-Cure on 1,056,739 reviews of 70 popular Android apps. Our evaluation demonstrates that LLM-Cure significantly outperforms the state-of-the-art approaches in assigning features to reviews by up to 13% in F1-score, up to 16% in recall and up to 11% in precision. Additionally, LLM-Cure demonstrates its capability to provide suggestions for resolving user complaints. We verify the suggestions using the release notes that reflect the changes of features in the target mobile app. LLM-Cure achieves a promising average of 73% of the implementation of the provided suggestions.

Taher A. Ghaleb, Daniel Alencar da Costa, Ying Zou

Continuous Integration (CI) provides early feedback by automatically building software, but long build durations can hinder developer productivity. CI services use caching to speed up builds by reusing infrequently changing artifacts, yet little is known about how caching is adopted in practice and what challenges it entails. In this paper, we conduct a large-scale empirical study of CI caching in Travis CI, analyzing 513,384 builds from 1,279 GitHub projects. We find that only 30% of projects adopt CI caching, and early adopters are typically more mature, with more dependencies, commits, and longer CI lifespans. To understand non-adoption, we submit pull requests enabling caching in non-adopting projects, and nearly half are accepted or merged. Developer feedback indicates that non- or late adoption mainly results from limited awareness of CI caching support. We further study cache maintenance and identify five common activities, performed by 24% of cache-enabled projects. While one-third of projects see substantial build-time reductions, cache uploads occur in 97% of builds, and 27% of projects contain stale cached artifacts. An analysis of reported caching issues shows developers mainly struggle with corrupted or outdated caches and request broader caching features. Overall, CI caching does not benefit all projects, requires ongoing maintenance, and is more complex in practice than many developers expect.

Fangjian Lei, Mariam El Mezouar, Shayan Noei et al.

Large Language Models (LLMs) have shown promise in assisting developers with code-related questions; however, LLMs carry the risk of generating unreliable answers. To address this, Retrieval-Augmented Generation (RAG) has been proposed to reduce the unreliability (i.e., hallucinations) of LLMs. However, designing effective pipelines remains challenging due to numerous design choices. In this paper, we construct a retrieval corpus of over 3 million Java and Python related Stack Overflow posts with accepted answers, and explore various RAG pipeline designs to answer developer questions, evaluating their effectiveness in generating accurate and reliable responses. More specifically, we (1) design and evaluate 7 different RAG pipelines and 63 pipeline variants to answer questions that have historically similar matches, and (2) address new questions without any close prior matches by automatically lowering the similarity threshold during retrieval, thereby increasing the chance of finding partially relevant context and improving coverage for unseen cases. We find that implementing a RAG pipeline combining hypothetical-documentation-embedding (HyDE) with the full-answer context performs best in retrieving and answering similarcontent for Stack Overflow questions. Finally, we apply our optimal RAG pipeline to 4 open-source LLMs and compare the results to their zero-shot performance. Our findings show that RAG with our optimal RAG pipeline consistently outperforms zero-shot baselines across models, achieving higher scores for helpfulness, correctness, and detail with LLM-as-a-judge. These findings demonstrate that our optimal RAG pipelines robustly enhance answer quality for a wide range of developer queries including both previously seen and novel questions across different LLMs

Jiayuan Zhou, Shaowei Wang, Cor-Paul Bezemer et al.

Due to the voluntary nature of open source software, it can be hard to find a developer to work on a particular task. For example, some issue reports may be too cumbersome and unexciting for someone to volunteer to do them, yet these issue reports may be of high priority to the success of a project. To provide an incentive for implementing such issue reports, one can propose a monetary reward, i.e., a bounty, to the developer who completes that particular task. In this paper, we study bounties in open source projects on GitHub to better understand how bounties can be leveraged to evolve such projects in terms of addressing issue reports. We investigated 5,445 bounties for GitHub projects. These bounties were proposed through the Bountysource platform with a total bounty value of $406,425. We find that 1) in general, the timing of proposing bounties and the bounty-usage frequency are the most important factors that impact the likelihood of an issue being addressed. More specifically, issue reports are more likely to be addressed if they are for projects in which bounties are used more frequently and if they are proposed earlier. 2) The bounty value that an issue report has is the most important factor that impacts the issue-addressing likelihood in the projects in which no bounties were used before. Backers in such projects proposed higher bounty values to get issues addressed. 3) There is a risk of wasting money for backers who invest money on long-standing issue reports.

Ming Lu, Zhen Gao, Ying Zou et al.

With the development of technology, the chemical production process is becoming increasingly complex and large-scale, making fault detection particularly important. However, current detective methods struggle to address the complexities of large-scale production processes. In this paper, we integrate the strengths of deep learning and machine learning technologies, combining the advantages of bidirectional long and short-term memory neural networks, fully connected neural networks, and the extra trees algorithm to propose a novel fault detection model named three-layer deep learning network random trees (TDLN-trees). First, the deep learning component extracts temporal features from industrial data, combining and transforming them into a higher-level data representation. Second, the machine learning component processes and classifies the features extracted in the first step. An experimental analysis based on the Tennessee Eastman process verifies the superiority of the proposed method.

Maram Assi, Safwat Hassan, Ying Zou

Deep Learning (DL) frameworks play a critical role in advancing artificial intelligence, and their rapid growth underscores the need for a comprehensive understanding of software quality and maintainability. DL frameworks, like other systems, are prone to code clones. Code clones refer to identical or highly similar source code fragments within the same project or even across different projects. Code cloning can have positive and negative implications for software development, influencing maintenance, readability, and bug propagation. In this paper, we aim to address the knowledge gap concerning the evolutionary dimension of code clones in DL frameworks and the extent of code reuse across these frameworks. We empirically analyze code clones in nine popular DL frameworks, i.e., TensorFlow, Paddle, PyTorch, Aesara, Ray, MXNet, Keras, Jax and BentoML, to investigate (1) the characteristics of the long-term code cloning evolution over releases in each framework, (2) the short-term, i.e., within-release, code cloning patterns and their influence on the long-term trends, and (3) the file-level code clones within the DL frameworks. Our findings reveal that DL frameworks adopt four distinct cloning trends and that these trends present some common and distinct characteristics. For instance, bug-fixing activities persistently happen in clones irrespective of the clone evolutionary trend but occur more in the "Serpentine" trend. Moreover, the within release level investigation demonstrates that short-term code cloning practices impact long-term cloning trends. The cross-framework code clone investigation reveals the presence of functional and architectural adaptation file-level cross-framework code clones across the nine studied frameworks. We provide insights that foster robust clone practices and collaborative maintenance in the development of DL frameworks.

Pouya Fathollahzadeh, Mariam El Mezouar, Hao Li et al.

In software engineering chatrooms, communication is often hindered by imprecise questions that cannot be answered. Recognizing key entities can be essential for improving question clarity and facilitating better exchange. However, existing research using natural language processing techniques often overlooks these software-specific nuances. In this paper, we introduce Software-specific Named Entity Recognition, Intent Detection, and Resolution Classification (SENIR), a labeling approach that leverages a Large Language Model to annotate entities, intents, and resolution status in developer chatroom conversations. To offer quantitative guidance for improving question clarity and resolvability, we build a resolution prediction model that leverages SENIR's entity and intent labels along with additional predictive features. We evaluate SENIR on the DISCO dataset using a subset of annotated chatroom dialogues. SENIR achieves an 86% F-score for entity recognition, a 71% F-score for intent detection, and an 89% F-score for resolution status classification. Furthermore, our resolution prediction model, tested with various sampling strategies (random undersampling and oversampling with SMOTE) and evaluation methods (5-fold cross-validation, 10-fold cross-validation, and bootstrapping), demonstrates AUC values ranging from 0.7 to 0.8. Key factors influencing resolution include positive sentiment and entities such as Programming Language and User Variable across multiple intents, while diagnostic entities are more relevant in error-related questions. Moreover, resolution rates vary significantly by intent: questions about API Usage and API Change achieve higher resolution rates, whereas Discrepancy and Review have lower resolution rates. A Chi-Square analysis confirms the statistical significance of these differences.

Yiping Jia, Zhen Ming Jiang, Shayan Noei et al.

With the widespread adoption of Large Language Models (LLMs) such as GitHub Copilot and ChatGPT, developers increasingly rely on AI-assisted tools to support code generation. While LLMs can generate syntactically correct solutions for well-structured programming tasks, they often struggle with challenges that require iterative debugging, error handling, or adaptation to diverse problem structures. Existing approaches such as fine-tuning or self-repair strategies either require costly retraining or lack mechanisms to accumulate and reuse knowledge from previous attempts. To address these limitations, we propose MemoCoder, a multi-agent framework that enables collaborative problem solving and persistent learning from past fixes. At the core of MemoCoder is a Fixing Knowledge Set, which stores successful repairs and supports retrieval for future tasks. A central Mentor Agent supervises the repair process by identifying recurring error patterns and refining high-level fixing strategies, providing a novel supervisory role that guides the self-repair loop. We evaluate MemoCoder across three public benchmarks -- MBPP, HumanEval, and LiveCodeBench -- spanning a range of problem complexities. Experimental results show that MemoCoder consistently outperforms both zero-shot prompting and a Self-Repair strategy, with improvements ranging from 3.1% to 12.1% in Pass@10 and from 1.4% to 14.5% in Pass@50, demonstrating its effectiveness in iterative refinement and knowledge-guided code generation.

Jiakun Liu, Haoxiang Zhang, Xin Xia et al.

On Stack Overflow, users reuse 11,926,354 external links to share the resources hosted outside the Stack Overflow website. The external links connect to the existing programming-related knowledge and extend the crowdsourced knowledge on Stack Overflow. Some of the external links, so-called as repeated external links, can be shared for multiple times. We observe that 82.5% of the link sharing activities (i.e., sharing links in any question, answer, or comment) on Stack Overflow share external resources, and 57.0% of the occurrences of the external links are sharing the repeated external links. However, it is still unclear what types of external resources are repeatedly shared. To help users manage their knowledge, we wish to investigate the characteristics of the repeated external links in knowledge sharing on Stack Overflow. In this paper, we analyze the repeated external links on Stack Overflow. We observe that external links that point to the text resources (hosted in documentation websites, tutorial websites, etc.) are repeatedly shared the most. We observe that: 1) different users repeatedly share the same knowledge in the form of repeated external links, thus increasing the maintenance effort of knowledge (e.g., update invalid links in multiple posts), 2) the same users can repeatedly share the external links for the purpose of promotion, and 3) external links can point to webpages with an overload of information that is difficult for users to retrieve relevant information. Our findings provide insights to Stack Overflow moderators and researchers. For example, we encourage Stack Overflow to centrally manage the commonly occurring knowledge in the form of repeated external links in order to better maintain the crowdsourced knowledge on Stack Overflow.

Jiakun Liu, Xin Xia, David Lo et al.

Stack Overflow hosts valuable programming-related knowledge with 11,926,354 links that reference to the third-party websites. The links that reference to the resources hosted outside the Stack Overflow websites extend the Stack Overflow knowledge base substantially. However, with the rapid development of programming-related knowledge, many resources hosted on the Internet are not available anymore. Based on our analysis of the Stack Overflow data that was released on Jun. 2, 2019, 14.2% of the links on Stack Overflow are broken links. The broken links on Stack Overflow can obstruct viewers from obtaining desired programming-related knowledge, and potentially damage the reputation of the Stack Overflow as viewers might regard the posts with broken links as obsolete. In this paper, we characterize the broken links on Stack Overflow. 65% of the broken links in our sampled questions are used to show examples, e.g., code examples. 70% of the broken links in our sampled answers are used to provide supporting information, e.g., explaining a certain concept and describing a step to solve a problem. Only 1.67% of the posts with broken links are highlighted as such by viewers in the posts' comments. Only 5.8% of the posts with broken links removed the broken links. Viewers cannot fully rely on the vote scores to detect broken links, as broken links are common across posts with different vote scores. The websites that host resources that can be maintained by their users are referenced by broken links the most on Stack Overflow -- a prominent example of such websites is GitHub. The posts and comments related to the web technologies, i.e., JavaScript, HTML, CSS, and jQuery, are associated with more broken links. Based on our findings, we shed lights for future directions and provide recommendations for practitioners and researchers.

Haoxiang Zhang, Shaowei Wang, Tse-Hsun et al.

Stack Overflow accumulates an enormous amount of software engineering knowledge. However, as time passes, certain knowledge in answers may become obsolete. Such obsolete answers, if not identified or documented clearly, may mislead answer seekers and cause unexpected problems (e.g., using an out-dated security protocol). In this paper, we investigate how the knowledge in answers becomes obsolete and identify the characteristics of such obsolete answers. We find that: 1) More than half of the obsolete answers (58.4%) were probably already obsolete when they were first posted. 2) When an obsolete answer is observed, only a small proportion (20.5%) of such answers are ever updated. 3) Answers to questions in certain tags (e.g., node.js, ajax, android, and objective-c) are more likely to become obsolete. Our findings suggest that Stack Overflow should develop mechanisms to encourage the whole community to maintain answers (to avoid obsolete answers) and answer seekers are encouraged to carefully go through all information (e.g., comments) in answer threads.