arXiv subjects

Yan Lei

Publications and source records attributed to Yan Lei.

13 recordsLinked to original sources

arXiv2026

PlanningBench: Generating Scalable and Verifiable Planning Data for Evaluating and Training Large Language Models

Planning is a fundamental capability for large language models (LLMs) because such complex tasks require models to coordinate goals, constraints, resources, and long-term consequences into executable and verifiable solutions. Existing planning benchmarks, however, usually treat planning data as fixed collections of instances rather than controllable generation targets. This limits scenario coverage, ties difficulty to surface-level proxies rather than structural sources, and offers limited support for scalable generation, automatic verification, or planning-oriented training. We introduce PlanningBench, a framework for generating scalable, diverse, and verifiable planning data for both evaluation and training. PlanningBench starts from real planning scenarios and abstracts practical workflows into a structured taxonomy of more than 30 task types, subtasks, constraint families, and difficulty factors. Guided by this taxonomy, a constraint-driven synthesis pipeline instantiates self-contained planning problems with adaptive difficulty control, quality filtering, and instance-level verification checklists. This shifts planning data construction from fixed benchmark collection to controllable generation while preserving realistic task grounding. We use PlanningBench to evaluate open-source and closed-source frontier LLMs, and find that current models still struggle to produce complete solutions under coupled constraints. Beyond evaluation, reinforcement learning on verified PlanningBench data improves performance on unseen planning benchmarks and broader instruction-following tasks. Further analysis suggests that determinate or well-specified optimal solutions provide clearer reward signals and more stable training dynamics. Overall, PlanningBench provides a controllable source of planning data for diagnosing and improving generalizable planning abilities in LLMs.

cs.AI↗

arXiv2026

Debug Like a Human: Scaling LLM-based Fault Localization to Processor Design via Block-Level Instruction-Oriented Slicing

Fault localization in modern processor design code is a critical yet time-consuming step during processor verification. While recent advances in LLM-based techniques for module-level hardware design have shown promising results, automatically localizing bugs in large-scale, project-level processor designs remains challenging. In this paper, we present BluesFL, a novel block-level LLM-based fault localization framework for processor designs. Inspired by the way engineers debug processors, we first propose a dataflow-based code blockization approach to guide LLMs to focus on critical local code context. We further propose a Block-Level Instruction-Oriented Slicing (Blues) algorithm that enables LLMs to mimic human reasoning by analyzing instruction execution paths and processor states. We evaluate BluesFL on a real-world RISC-V processor core comprising 19K lines of SystemVerilog code. Experimental results demonstrate that BluesFL correctly localizes 24 bugs at Top-1, achieving 242.9% improvement over the existing state-of-the-art (7 bugs). Cost analysis shows that BluesFL requires an average of only $0.257 to localize a single bug.

cs.SE↗

arXiv2026

CL-bench Life: Can Language Models Learn from Real-Life Context?

Today's AI assistants such as OpenClaw are designed to handle context effectively, making context learning an increasingly important capability for models. As these systems move beyond professional settings into everyday life, the nature of the contexts they must handle also shifts. Real-life contexts are often messy, fragmented, and deeply tied to personal and social experience, such as multi-party conversations, personal archives, and behavioral traces. Yet it remains unclear whether current frontier language models can reliably learn from such contexts and solve tasks grounded in them. To this end, we introduce CL-bench Life, a fully human-curated benchmark comprising 405 context-task pairs and 5,348 verification rubrics, covering common real-life scenarios. Solving tasks in CL-bench Life requires models to reason over complex, messy real-life contexts, calling for strong real-life context learning abilities that go far beyond those evaluated in existing benchmarks. We evaluate ten frontier LMs and find that real-life context learning remains highly challenging: even the best-performing model achieves only 19.3% task solving rate, while the average performance across models is only 13.8%. Models still struggle to reason over contexts such as messy group chat histories and fragmented behavioral records from everyday life. CL-bench Life provides a crucial testbed for advancing real-life context learning, and progress on it can enable more intelligent and reliable AI assistants in everyday life.

cs.CL↗

arXiv2025

DynaFix: Iterative Automated Program Repair Driven by Execution-Level Dynamic Information

Automated Program Repair (APR) aims to automatically generate correct patches for buggy programs. Recent approaches leveraging large language models (LLMs) have shown promise but face limitations. Most rely solely on static analysis, ignoring runtime behaviors. Some attempt to incorporate dynamic signals, but these are often restricted to training or fine-tuning, or injected only once into the repair prompt, without iterative use. This fails to fully capture program execution. Current iterative repair frameworks typically rely on coarse-grained feedback, such as pass/fail results or exception types, and do not leverage fine-grained execution-level information effectively. As a result, models struggle to simulate human stepwise debugging, limiting their effectiveness in multi-step reasoning and complex bug repair. To address these challenges, we propose DynaFix, an execution-level dynamic information-driven APR method that iteratively leverages runtime information to refine the repair process. In each repair round, DynaFix captures execution-level dynamic information such as variable states, control-flow paths, and call stacks, transforming them into structured prompts to guide LLMs in generating candidate patches. If a patch fails validation, DynaFix re-executes the modified program to collect new execution information for the next attempt. This iterative loop incrementally improves patches based on updated feedback, similar to the stepwise debugging practices of human developers. We evaluate DynaFix on the Defects4J v1.2 and v2.0 benchmarks. DynaFix repairs 186 single-function bugs, a 10% improvement over state-of-the-art baselines, including 38 bugs previously unrepaired. It achieves correct patches within at most 35 attempts, reducing the patch search space by 70% compared with existing methods, thereby demonstrating both effectiveness and efficiency in repairing complex bugs.

cs.SE↗

arXiv2025

ZoFia: Zero-Shot Fake News Detection with Entity-Guided Retrieval and Multi-LLM Interaction

The rapid spread of fake news threatens social stability and public trust, highlighting the urgent need for its effective detection. Although large language models (LLMs) show potential in fake news detection, they are limited by knowledge cutoff and easily generate factual hallucinations when handling time-sensitive news. Furthermore, the thinking of a single LLM easily falls into early stance locking and confirmation bias, making it hard to handle both content reasoning and fact checking simultaneously. To address these challenges, we propose ZoFia, a two-stage zero-shot fake news detection framework. In the first retrieval stage, we propose novel Hierarchical Salience and Salience-Calibrated Minimum Marginal Relevance (SC-MMR) algorithm to extract core entities accurately, which drive dual-source retrieval to overcome knowledge and evidence gaps. In the subsequent stage, a multi-agent system conducts multi-perspective reasoning and verification in parallel and achieves an explainable and robust result via adversarial debate. Comprehensive experiments on two public datasets show that ZoFia outperforms existing zero-shot baselines and even most few-shot methods. Our code has been open-sourced to facilitate the research community at https://github.com/SakiRinn/ZoFia.

cs.CL↗

arXiv2025

Principal Context-aware Diffusion Guided Data Augmentation for Fault Localization

Test cases are indispensable for conducting effective fault localization (FL). However, test cases in practice are severely class imbalanced, i.e. the number of failing test cases (i.e. minority class) is much less than that of passing ones (i.e. majority class). The severe class imbalance between failing and passing test cases have hindered the FL effectiveness. To address this issue, we propose PCD-DAug: a Principal Context-aware Diffusion guided Data Augmentation approach that generate synthesized failing test cases for improving FL. PCD-DAug first combines program slicing with principal component analysis to construct a principal context that shows how a set of statements influences the faulty output via statistical program dependencies. Then, PCD-DAug devises a conditional diffusion model to learn from principle contexts for generating synthesized failing test cases and acquiring a class balanced dataset for FL. We conducted large-scale experiments on six state-of-the-art FL approaches and compare PCD-DAug with six data augmentation baselines. The results show that PCD-DAug significantly improves FL effectiveness, e.g. achieving average improvements of 383.83%, 227.08%, and 224.19% in six FL approaches under the metrics Top-1, Top-3, and Top-5, respectively.

cs.SE↗

arXiv2024

Fault Localization from the Semantic Code Search Perspective

The software development process is characterized by an iterative cycle of continuous functionality implementation and debugging, essential for the enhancement of software quality and adaptability to changing requirements. This process incorporates two isolatedly studied tasks: Code Search (CS), which retrieves reference code from a code corpus to aid in code implementation, and Fault Localization (FL), which identifies code entities responsible for bugs within the software project to boost software debugging. These two tasks exhibit similarities since they both address search problems. Notably, CS techniques have demonstrated greater effectiveness than FL ones, possibly because of the precise semantic details of the required code offered by natural language queries, which are not readily accessible to FL methods. Drawing inspiration from this, we hypothesize that a fault localizer could achieve greater proficiency if semantic information about the buggy methods were made available. Based on this idea, we propose CosFL, an FL approach that decomposes the FL task into two steps: query generation, which describes the functionality of the problematic code in natural language, and fault retrieval, which uses CS to find program elements semantically related to the query. Specifically, to depict the buggy functionalities and generate high-quality queries, CosFL extensively harnesses the code analysis, semantic comprehension, and decision-making capabilities of LLMs. Moreover, to enhance the accuracy of CS, CosFL captures varying levels of context information and employs a multi-granularity code search strategy, which facilitates a more precise identification of buggy methods from a holistic view. The evaluation on 835 real bugs from 23 Java projects shows that CosFL successfully localizes 324 bugs within Top-1, which significantly outperforms the state-of-the-art approaches by 26.6%-57.3%.

cs.SE↗

arXiv2024

Taking a Deep Breath: Enhancing Language Modeling of Large Language Models with Sentinel Tokens

Large language models (LLMs) have shown promising efficacy across various tasks, becoming powerful tools in numerous aspects of human life. However, Transformer-based LLMs suffer a performance degradation when modeling long-term contexts due to they discard some information to reduce computational overhead. In this work, we propose a simple yet effective method to enable LLMs to take a deep breath, encouraging them to summarize information contained within discrete text chunks. Specifically, we segment the text into multiple chunks and insert special token at the end of each chunk. We then modify the attention mask to integrate the chunk's information into the corresponding token. This facilitates LLMs to interpret information not only from historical individual tokens but also from the token, aggregating the chunk's semantic information. Experiments on language modeling and out-of-domain downstream tasks validate the superiority of our approach.

cs.CL↗

arXiv2024

Qsnail: A Questionnaire Dataset for Sequential Question Generation

The questionnaire is a professional research methodology used for both qualitative and quantitative analysis of human opinions, preferences, attitudes, and behaviors. However, designing and evaluating questionnaires demands significant effort due to their intricate and complex structure. Questionnaires entail a series of questions that must conform to intricate constraints involving the questions, options, and overall structure. Specifically, the questions should be relevant and specific to the given research topic and intent. The options should be tailored to the questions, ensuring they are mutually exclusive, completed, and ordered sensibly. Moreover, the sequence of questions should follow a logical order, grouping similar topics together. As a result, automatically generating questionnaires presents a significant challenge and this area has received limited attention primarily due to the scarcity of high-quality datasets. To address these issues, we present Qsnail, the first dataset specifically constructed for the questionnaire generation task, which comprises 13,168 human-written questionnaires gathered from online platforms. We further conduct experiments on Qsnail, and the results reveal that retrieval models and traditional generative models do not fully align with the given research topic and intents. Large language models, while more closely related to the research topic and intents, exhibit significant limitations in terms of diversity and specificity. Despite enhancements through the chain-of-thought prompt and finetuning, questionnaires generated by language models still fall short of human-written questionnaires. Therefore, questionnaire generation is challenging and needs to be further explored. The dataset is available at: https://github.com/LeiyanGithub/qsnail.

cs.CL↗

arXiv2023

Revisiting and Improving Retrieval-Augmented Deep Assertion Generation

Unit testing validates the correctness of the unit under test and has become an essential activity in software development process. A unit test consists of a test prefix that drives the unit under test into a particular state, and a test oracle (e.g., assertion), which specifies the behavior in that state. To reduce manual efforts in conducting unit testing, Yu et al. proposed an integrated approach (integration for short), combining information retrieval (IR) with a deep learning-based approach, to generate assertions for a unit test. Despite promising, there is still a knowledge gap as to why or where integration works or does not work. In this paper, we describe an in-depth analysis of the effectiveness of integration. Our analysis shows that: 1) The overall performance of integration is mainly due to its success in retrieving assertions. 2) integration struggles to understand the semantic differences between the retrieved focal-test (focal-test includes a test prefix and a unit under test) and the input focal-test; 3) integration is limited to specific types of edit operations and cannot handle token addition or deletion. To improve the effectiveness of assertion generation, this paper proposes a novel retrieve-and-edit approach named EditAS. Specifically, EditAS first retrieves a similar focal-test from a pre-defined corpus and treats its assertion as a prototype. Then, EditAS reuses the information in the prototype and edits the prototype automatically. EditAS is more generalizable than integration. We conduct experiments on two large-scale datasets and experimental results demonstrate that EditAS outperforms the state-of-the-art approaches, with an average improvement of 10.00%-87.48% and 3.30%-42.65% in accuracy and BLEU score, respectively.

cs.SE↗

arXiv2023

Mitigating the Effect of Class Imbalance in Fault Localization Using Context-aware Generative Adversarial Network

Fault localization (FL) analyzes the execution information of a test suite to pinpoint the root cause of a failure. The class imbalance of a test suite, i.e., the imbalanced class proportion between passing test cases (i.e., majority class) and failing ones (i.e., minority class), adversely affects FL effectiveness. To mitigate the effect of class imbalance in FL, we propose CGAN4FL: a data augmentation approach using Context-aware Generative Adversarial Network for Fault Localization. Specifically, CGAN4FL uses program dependencies to construct a failure-inducing context showing how a failure is caused. Then, CGAN4FL leverages a generative adversarial network to analyze the failure-inducing context and synthesize the minority class of test cases (i.e., failing test cases). Finally, CGAN4FL augments the synthesized data into original test cases to acquire a class-balanced dataset for FL. Our experiments show that CGAN4FL significantly improves FL effectiveness, e.g., promoting MLP-FL by 200.00%, 25.49%, and 17.81% under the Top-1, Top-5, and Top-10 respectively.

cs.SE↗

arXiv2020

Deep Double-Side Learning Ensemble Model for Few-Shot Parkinson Speech Recognition

Diagnosis and therapeutic effect assessment of Parkinson disease based on voice data are very important,but its few-shot learning problem is challenging.Although deep learning is good at automatic feature extraction, it suffers from few-shot learning problem. Therefore, the general effective method is first conduct feature extraction based on prior knowledge, and then carry out feature reduction for subsequent classification. However, there are two major problems: 1) Structural information among speech features has not been mined and new features of higher quality have not been reconstructed. 2) Structural information between data samples has not been mined and new samples with higher quality have not been reconstructed. To solve these two problems, based on the existing Parkinson speech feature data set, a deep double-side learning ensemble model is designed in this paper that can reconstruct speech features and samples deeply and simultaneously. As to feature reconstruction, an embedded deep stacked group sparse auto-encoder is designed in this paper to conduct nonlinear feature transformation, so as to acquire new high-level deep features, and then the deep features are fused with original speech features by L1 regularization feature selection method. As to speech sample reconstruction, a deep sample learning algorithm is designed in this paper based on iterative mean clustering to conduct samples transformation, so as to obtain new high-level deep samples. Finally, the bagging ensemble learning mode is adopted to fuse the deep feature learning algorithm and the deep samples learning algorithm together, thereby constructing a deep double-side learning ensemble model. At the end of this paper, two representative speech datasets of Parkinson's disease were used for verification. The experimental results show that the proposed algorithm are effective.

cs.CV↗

arXiv2020

Hybrid Embedded Deep Stacked Sparse Autoencoder with w_LPPD SVM Ensemble

Deep learning is a kind of feature learning method with strong nonliear feature transformation and becomes more and more important in many fields of artificial intelligence. Deep autoencoder is one representative method of the deep learning methods, and can effectively extract abstract the information of datasets. However, it does not consider the complementarity between the deep features and original features during deep feature transformation. Besides, it suffers from small sample problem. In order to solve these problems, a novel deep autoencoder - hybrid feature embedded stacked sparse autoencoder(HESSAE) has been proposed in this paper. HFESAE is capable to learn discriminant deep features with the help of embedding original features to filter weak hidden-layer outputs during training. For the issue that class representation ability of abstract information is limited by small sample problem, a feature fusion strategy has been designed aiming to combining abstract information learned by HFESAE with original feature and obtain hybrid features for feature reduction. The strategy is hybrid feature selection strategy based on L1 regularization followed by an support vector machine(SVM) ensemble model, in which weighted local discriminant preservation projection (w_LPPD), is designed and employed on each base classifier. At the end of this paper, several representative public datasets are used to verify the effectiveness of the proposed algorithm. The experimental results demonstrated that, the proposed feature learning method yields superior performance compared to other existing and state of art feature learning algorithms including some representative deep autoencoder methods.

cs.LG↗