arXiv subjects

Neng Zhang

Publications and source records attributed to Neng Zhang.

13 recordsLinked to original sources

arXiv2026

Diffusion Forcing Planner: History-Annealed Planning with Time-Dependent Guidance for Autonomous Driving

Learning-based motion planners, despite recent progress, often suffer from temporal inconsistency. Small perturbations across frames can accumulate into unstable trajectories, degrading comfort and safety in closed-loop driving. Several methods attempt to inject history as a static conditioning signal to stabilize outputs, only to induce the planner to copy historical patterns instead of adapting to environment contexts. To address this limitation, we propose Diffusion Forcing Planner (DFP), a diffusion-based planning framework driven by history-guided control. Specifically, DFP decomposes the full trajectory into history, current and future segments, and assign independent noise levels to each segment. The model jointly denoises the historical and the future segments, enforcing a heterogeneous joint diffusion process. At inference, classifier-free guidance (CFG) is applied to steer future sampling using annealed history in a controllable manner. Closed-loop evaluation and comprehensive ablations on nuPlan show that DFP achieves competitive performance while producing continuous, stable, and controllable motion plans in complex driving scenarios.

cs.RO↗

arXiv2026

TICoder: A Repository-Level Code Generation Framework with Test-Driven Planning and Implementation-Aware Reuse

Repository-level code generation with Large Language Models (LLMs) remains challenging, primarily due to complex dependencies and limited context windows. Recent approaches adopt retrieval-augmented generation (RAG) and the planning mechanism to reuse potential callee functions in the repository. However, these approaches often suffer from two limitations: lack of test-driven behavioral guidance during planning and overlooking the implementation logic embedded in repository code during reuse. As a result, generated plans may not align with expected behaviors, and retrieved functions may not be effectively reused. In this paper, we propose TICoder, a novel repository-level code generation framework that improves both planning and reuse. TICoder introduces a test-driven iterative planning mechanism that leverages test cases as behavioral specifications to refine implementation steps. Furthermore, TICoder employs an implementation-aware code reuse strategy, which retrieves potential callee functions using a dual-view similarity that captures both functional and implementation aspects. We then identify relevant usage patterns through a dual-stage selection strategy, combining structure-based clustering and perplexity-based filtering. We conduct extensive experiments on widely used repository-level code generation benchmarks with various LLMs. Experimental results demonstrate that TICoder outperforms state-of-the-art (SOTA) methods, achieving an average improvement of 11.52%.

cs.SE↗

arXiv2026

From Language to Action: Enhancing LLM Task Efficiency with Task-Aware MCP Server Recommendation

The rapid expansion of the model context protocol (MCP) ecosystem enables large language model (LLM)-based agents to access a wide range of external tools via a standardized interface. However, identifying appropriate MCP servers for a specific development task remains challenging. Existing studies primarily focus on measuring the MCP ecosystem or optimizing tool invocation mechanisms, while systematic recommendation frameworks and reproducible benchmarks for real-world development tasks remain largely unexplored. To address this limitation, we formulate task-oriented MCP server recommendation as a structured retrieval-and-ranking problem that jointly considers semantic relevance and engineering constraints. We first construct Task2MCP, a task-centered dataset that systematically associates taxonomy-grounded development tasks with curated MCP servers. This dataset provides structured supervision and a reproducible evaluation environment for research on MCP tool recommendations. Building on this dataset, we propose T2MRec, a task-to-MCP server recommendation model. It models semantic relevance and structural compatibility to construct an initial candidate set. Then it improves coverage and ranking quality through centroid-based candidate expansion and constrained LLM-based re-ranking. In addition, we design and implement an interactive MCP server recommendation agent prototype that operates in conversational environments to support dynamic decision-making. The agent assists developers in efficiently evaluating and integrating tools by providing recommended MCP servers together with usage guidelines.

cs.SE↗

arXiv2026

FastFI: Enhancing API Call-Site Robustness in Microservice-Based Systems with Fault Injection

Fault injection is a key technique for assessing software reliability, enabling proactive detection of system defects before they manifest in production. However, the increasing complexity of microservice architectures leads to exponential growth in the fault-injection space, rendering traditional random injection inefficient. Recent lineage-driven approaches mitigate this problem through heuristic pruning, but they face two limitations. First, combinatorial-fault discovery remains bottlenecked by general-purpose SAT solvers, which fail to exploit the monotone and low-overlap structure of derived CNF formulas and typically rely on a static upper bound on fault size. Second, existing techniques provide limited post-injection guidance beyond reporting detected faults. To address these challenges, we propose FastFI, a fault-injection-guided framework to enhance the robustness of API call sites in microservice-based systems. FastFI features a DFS-based solver with dynamic fault injection to discover all valid combinatorial faults, and it leverages fault-injection results to identify critical APIs whose call sites should be hardened for robustness. Experiments on four representative microservice benchmarks show that FastFI reduces end-to-end fault-injection time by an average of 76.12\% compared to state-of-the-art baselines while maintaining acceptable resource overhead. Moreover, FastFI accurately identifies high-impact APIs and provides actionable guidance for call-site hardening.

cs.SE↗

arXiv2025

WAM-Diff: A Masked Diffusion VLA Framework with MoE and Online Reinforcement Learning for Autonomous Driving

End-to-end autonomous driving systems based on vision-language-action (VLA) models integrate multimodal sensor inputs and language instructions to generate planning and control signals. While autoregressive large language models and continuous diffusion policies are prevalent, the potential of discrete masked diffusion for trajectory generation remains largely unexplored. This paper presents WAM-Diff, a VLA framework that employs masked diffusion to iteratively refine a discrete sequence representing future ego-trajectories. Our approach features three key innovations: a systematic adaptation of masked diffusion for autonomous driving that supports flexible, non-causal decoding orders; scalable model capacity via a sparse MoE architecture trained jointly on motion prediction and driving-oriented visual question answering (VQA); and online reinforcement learning using Group Sequence Policy Optimization (GSPO) to optimize sequence-level driving rewards. Remarkably, our model achieves 91.0 PDMS on NAVSIM-v1 and 89.7 EPDMS on NAVSIM-v2, demonstrating the effectiveness of masked diffusion for autonomous driving. The approach provides a promising alternative to autoregressive and diffusion-based policies, supporting scenario-aware decoding strategies for trajectory generation. The code for this paper will be released publicly at: https://github.com/fudan-generative-vision/WAM-Diff

cs.RO↗

arXiv2025

WAM-Flow: Parallel Coarse-to-Fine Motion Planning via Discrete Flow Matching for Autonomous Driving

We introduce WAM-Flow, a vision-language-action (VLA) model that casts ego-trajectory planning as discrete flow matching over a structured token space. In contrast to autoregressive decoders, WAM-Flow performs fully parallel, bidirectional denoising, enabling coarse-to-fine refinement with a tunable compute-accuracy trade-off. Specifically, the approach combines a metric-aligned numerical tokenizer that preserves scalar geometry via triplet-margin learning, a geometry-aware flow objective and a simulator-guided GRPO alignment that integrates safety, ego progress, and comfort rewards while retaining parallel generation. A multi-stage adaptation converts a pre-trained auto-regressive backbone (Janus-1.5B) from causal decoding to non-causal flow model and strengthens road-scene competence through continued multimodal pretraining. Thanks to the inherent nature of consistency model training and parallel decoding inference, WAM-Flow achieves superior closed-loop performance against autoregressive and diffusion-based VLA baselines, with 1-step inference attaining 89.1 PDMS and 5-step inference reaching 90.3 PDMS on NAVSIM v1 benchmark. These results establish discrete flow matching as a new promising paradigm for end-to-end autonomous driving. The code will be publicly available soon.

cs.RO↗

arXiv2024

TVDiag: A Task-oriented and View-invariant Failure Diagnosis Framework with Multimodal Data

Microservice-based systems often suffer from reliability issues due to their intricate interactions and expanding scale. With the rapid growth of observability techniques, various methods have been proposed to achieve failure diagnosis, including root cause localization and failure type identification, by leveraging diverse monitoring data such as logs, metrics, or traces. However, traditional failure diagnosis methods that use single-modal data can hardly cover all failure scenarios due to the restricted information. Several failure diagnosis methods have been recently proposed to integrate multimodal data based on deep learning. These methods, however, tend to combine modalities indiscriminately and treat them equally in failure diagnosis, ignoring the relationship between specific modalities and different diagnostic tasks. This oversight hinders the effective utilization of the unique advantages offered by each modality. To address the limitation, we propose \textit{TVDiag}, a multimodal failure diagnosis framework for locating culprit microservice instances and identifying their failure types (e.g., Net-packets Corruption) in microservice-based systems. \textit{TVDiag} employs task-oriented learning to enhance the potential advantages of each modality and establishes cross-modal associations based on contrastive learning to extract view-invariant failure information. Furthermore, we develop a graph-level data augmentation strategy that randomly inactivates the observability of some normal microservice instances during training to mitigate the shortage of training data. Experimental results show that \textit{TVDiag} outperforms state-of-the-art methods in multimodal failure diagnosis, achieving at least a 55.94\% higher $HR@1$ accuracy and over a 4.08\% increase in F1-score across two datasets.

cs.SE↗

arXiv2024

Root Cause Localization for Microservice Systems in Cloud-edge Collaborative Environments

With the development of cloud-native technologies, microservice-based software systems face challenges in accurately localizing root causes when failures occur. Additionally, the cloud-edge collaborative environment introduces more difficulties, such as unstable networks and high latency across network segments. Accurately identifying the root cause of microservices in a cloud-edge collaborative environment has thus become an urgent problem. In this paper, we propose MicroCERCL, a novel approach that pinpoints root causes at the kernel and application level in the cloud-edge collaborative environment. Our key insight is that failures propagate through direct invocations and indirect resource-competition dependencies in a cloud-edge collaborative environment characterized by instability and high latency. This will become more complex in the hybrid deployment that simultaneously involves multiple microservice systems. Leveraging this insight, we extract valid contents from kernel-level logs to prioritize localizing the kernel-level root cause. Moreover, we construct a heterogeneous dynamic topology stack and train a graph neural network model to accurately localize the application-level root cause without relying on historical data. Notably, we released the first benchmark hybrid deployment microservice system in a cloud-edge collaborative environment (the largest and most complex within our knowledge). Experiments conducted on the dataset collected from the benchmark show that MicroCERCL can accurately localize the root cause of microservice systems in such environments, significantly outperforming state-of-the-art approaches with an increase of at least 24.1% in top-1 accuracy.

cs.SE↗

arXiv2024

An Empirical Study of ChatGPT-Related Projects and Their Issues on GitHub

Since the launch of ChatGPT in 2022, an increasing number of ChatGPT-related projects are being published on GitHub, sparking widespread discussions. However, GitHub does not provide a detailed classification of these projects to help users effectively explore interested projects. Additionally, the issues raised by users for these projects cover various aspects, e.g., installation, usage, and updates. It would be valuable to help developers prioritize more urgent issues and improve development efficiency. We retrieved 71,244 projects from GitHub using the keyword `ChatGPT' and selected the top 200 representative projects with the highest numbers of stars as our dataset. By analyzing the project descriptions, we identified three primary categories of ChatGPT-related projects, namely ChatGPT Implementation & Training, ChatGPT Application, ChatGPT Improvement & Extension. Next, we applied a topic modeling technique to 23,609 issues of those projects and identified ten issue topics, e.g., model reply and interaction interface. We further analyzed the popularity, difficulty, and evolution of each issue topic within the three project categories. Our main findings are: 1) The increase in the number of projects within the three categories is closely related to the development of ChatGPT; and 2) There are significant differences in the popularity, difficulty, and evolutionary trends of the issue topics across the three project categories. Based on these findings, we finally provided implications for project developers and platform managers on how to better develop and manage ChatGPT-related projects.

cs.SE↗

arXiv2024

iJTyper: An Iterative Type Inference Framework for Java by Integrating Constraint- and Statistically-based Methods

Inferring the types of API elements in incomplete code snippets (e.g., those on Q&A forums) is a prepositive step required to work with the code snippets. Existing type inference methods can be mainly categorized as constraint-based or statistically-based. The former imposes higher requirements on code syntax and often suffers from low recall due to the syntactic limitation of code snippets. The latter relies on the statistical regularities learned from a training corpus and does not take full advantage of the type constraints in code snippets, which may lead to low precision. In this paper, we propose an iterative type inference framework for Java, called iJTyper, by integrating the strengths of both constraint- and statistically-based methods. For a code snippet, iJTyper first applies a constraint-based method and augments the code context with the inferred types of API elements. iJTyper then applies a statistically-based method to the augmented code snippet. The predicted candidate types of API elements are further used to improve the constraint-based method by reducing its pre-built knowledge base. iJTyper iteratively executes both methods and performs code context augmentation and knowledge base reduction until a termination condition is satisfied. Finally, the final inference results are obtained by combining the results of both methods. We evaluated iJTyper on two open-source datasets. Results show that 1) iJTyper achieves high average precision/recall of 97.31% and 92.52% on both datasets; 2) iJTyper significantly improves the recall of two state-of-the-art baselines, SnR and MLMTyper, by at least 7.31% and 27.44%, respectively; and 3) iJTyper improves the average precision/recall of the popular language model, ChatGPT, by 3.25% and 0.51% on both datasets.

cs.SE↗

arXiv2023

Improving ChatGPT Prompt for Code Generation

Automated code generation can be a powerful technique for software development, significantly reducing developers' efforts and time required to create new code by generating it automatically based on requirements. Recently, OpenAI's language model ChatGPT has emerged as a powerful tool for generating human-like responses to a wide range of textual inputs (i.e., prompts), including those related to code generation. However, the effectiveness of ChatGPT for code generation is not well understood, and the generation performance could be heavily influenced by the choice of prompt. To answer these questions, we conducted experiments using the CodeXGlue dataset to evaluate ChatGPT's capabilities for two code generation tasks, including text-to-code and code-to-code generation. We designed prompts by leveraging the chain-of-thought strategy with multi-step optimizations. Our results showed that by carefully designing prompts to guide ChatGPT, the generation performance can be improved substantially. We also analyzed the factors that influenced the prompt design and provided insights that could guide future research.

cs.SE↗

arXiv2023

Turn the Rudder: A Beacon of Reentrancy Detection for Smart Contracts on Ethereum

Smart contracts are programs deployed on a blockchain and are immutable once deployed. Reentrancy, one of the most important vulnerabilities in smart contracts, has caused millions of dollars in financial loss. Many reentrancy detection approaches have been proposed. It is necessary to investigate the performance of these approaches to provide useful guidelines for their application. In this work, we conduct a large-scale empirical study on the capability of five well-known or recent reentrancy detection tools such as Mythril and Sailfish. We collect 230,548 verified smart contracts from Etherscan and use detection tools to analyze 139,424 contracts after deduplication, which results in 21,212 contracts with reentrancy issues. Then, we manually examine the defective functions located by the tools in the contracts. From the examination results, we obtain 34 true positive contracts with reentrancy and 21,178 false positive contracts without reentrancy. We also analyze the causes of the true and false positives. Finally, we evaluate the tools based on the two kinds of contracts. The results show that more than 99.8% of the reentrant contracts detected by the tools are false positives with eight types of causes, and the tools can only detect the reentrancy issues caused by call.value(), 58.8% of which can be revealed by the Ethereum's official IDE, Remix. Furthermore, we collect real-world reentrancy attacks reported in the past two years and find that the tools fail to find any issues in the corresponding contracts. Based on the findings, existing works on reentrancy detection appear to have very limited capability, and researchers should turn the rudder to discover and detect new reentrancy patterns except those related to call.value().

cs.SE↗

arXiv2021

Data privacy protection in microscopic image analysis for material data mining

Recent progress in material data mining has been driven by high-capacity models trained on large datasets. However, collecting experimental data has been extremely costly owing to the amount of human effort and expertise required. Therefore, material researchers are often reluctant to easily disclose their private data, which leads to the problem of data island, and it is difficult to collect a large amount of data to train high-quality models. In this study, a material microstructure image feature extraction algorithm FedTransfer based on data privacy protection is proposed. The core contributions are as follows: 1) the federated learning algorithm is introduced into the polycrystalline microstructure image segmentation task to make full use of different user data to carry out machine learning, break the data island and improve the model generalization ability under the condition of ensuring the privacy and security of user data; 2) A data sharing strategy based on style transfer is proposed. By sharing style information of images that is not urgent for user confidentiality, it can reduce the performance penalty caused by the distribution difference of data among different users.

eess.IV↗