arXiv subjects

Xin Xia

Publications and source records attributed to Xin Xia.

At least 91 records · Page 5Linked to original sources

arXiv2025

V2XPnP: Vehicle-to-Everything Spatio-Temporal Fusion for Multi-Agent Perception and Prediction

Vehicle-to-everything (V2X) technologies offer a promising paradigm to mitigate the limitations of constrained observability in single-vehicle systems. Prior work primarily focuses on single-frame cooperative perception, which fuses agents' information across different spatial locations but ignores temporal cues and temporal tasks (e.g., temporal perception and prediction). In this paper, we focus on the spatio-temporal fusion in V2X scenarios and design one-step and multi-step communication strategies (when to transmit) as well as examine their integration with three fusion strategies - early, late, and intermediate (what to transmit), providing comprehensive benchmarks with 11 fusion models (how to fuse). Furthermore, we propose V2XPnP, a novel intermediate fusion framework within one-step communication for end-to-end perception and prediction. Our framework employs a unified Transformer-based architecture to effectively model complex spatio-temporal relationships across multiple agents, frames, and high-definition maps. Moreover, we introduce the V2XPnP Sequential Dataset that supports all V2X collaboration modes and addresses the limitations of existing real-world datasets, which are restricted to single-frame or single-mode cooperation. Extensive experiments demonstrate that our framework outperforms state-of-the-art methods in both perception and prediction tasks.

cs.CV↗

arXiv2025

Refactoring Deep Learning Code: A Study of Practices and Unsatisfied Tool Needs

With the rapid development of deep learning, the implementation of intricate algorithms and substantial data processing have become standard elements of deep learning projects. As a result, the code has become progressively complex as the software evolves, which is difficult to maintain and understand. Existing studies have investigated the impact of refactoring on software quality within traditional software. However, the insight of code refactoring in the context of deep learning is still unclear. This study endeavors to fill this knowledge gap by empirically examining the current state of code refactoring in deep learning realm, and practitioners' views on refactoring. We first manually analyzed the commit history of five popular and well-maintained deep learning projects (e.g., PyTorch). We mined 4,921 refactoring practices in historical commits and measured how different types and elements of refactoring operations are distributed and found that refactoring operation types' distribution in deep learning projects is different from it in traditional Java software. We then surveyed 159 practitioners about their views of code refactoring in deep learning projects and their expectations of current refactoring tools. The result of the survey showed that refactoring research and the development of related tools in the field of deep learning are crucial for improving project maintainability and code quality, and that current refactoring tools do not adequately meet the needs of practitioners. Lastly, we provided our perspective on the future advancement of refactoring tools and offered suggestions for developers' development practices.

cs.SE↗

arXiv2025

CDA-SimBoost: A Unified Framework Bridging Real Data and Simulation for Infrastructure-Based CDA Systems

Cooperative Driving Automation (CDA) has garnered increasing research attention, yet the role of intelligent infrastructure remains insufficiently explored. Existing solutions offer limited support for addressing long-tail challenges, real-synthetic data fusion, and heterogeneous sensor management. This paper introduces CDA-SimBoost, a unified framework that constructs infrastructure-centric simulation environments from real-world data. CDA-SimBoost consists of three main components: a Digital Twin Builder for generating high-fidelity simulator assets based on sensor and HD map data, OFDataPip for processing both online and offline data streams, and OpenCDA-InfraX, a high-fidelity platform for infrastructure-focused simulation. The system supports realistic scenario construction, rare event synthesis, and scalable evaluation for CDA research. With its modular architecture and standardized benchmarking capabilities, CDA-SimBoost bridges real-world dynamics and virtual environments, facilitating reproducible and extensible infrastructure-driven CDA studies. All resources are publicly available at https://github.com/zhz03/CDA-SimBoost

eess.SY↗

arXiv2025

Adversarial Distribution Matching for Diffusion Distillation Towards Efficient Image and Video Synthesis

Distribution Matching Distillation (DMD) is a promising score distillation technique that compresses pre-trained teacher diffusion models into efficient one-step or multi-step student generators. Nevertheless, its reliance on the reverse Kullback-Leibler (KL) divergence minimization potentially induces mode collapse (or mode-seeking) in certain applications. To circumvent this inherent drawback, we propose Adversarial Distribution Matching (ADM), a novel framework that leverages diffusion-based discriminators to align the latent predictions between real and fake score estimators for score distillation in an adversarial manner. In the context of extremely challenging one-step distillation, we further improve the pre-trained generator by adversarial distillation with hybrid discriminators in both latent and pixel spaces. Different from the mean squared error used in DMD2 pre-training, our method incorporates the distributional loss on ODE pairs collected from the teacher model, and thus providing a better initialization for score distillation fine-tuning in the next stage. By combining the adversarial distillation pre-training with ADM fine-tuning into a unified pipeline termed DMDX, our proposed method achieves superior one-step performance on SDXL compared to DMD2 while consuming less GPU time. Additional experiments that apply multi-step ADM distillation on SD3-Medium, SD3.5-Large, and CogVideoX set a new benchmark towards efficient image and video synthesis.

cs.CV↗

arXiv2025

Seedream 3.0 Technical Report

We present Seedream 3.0, a high-performance Chinese-English bilingual image generation foundation model. We develop several technical improvements to address existing challenges in Seedream 2.0, including alignment with complicated prompts, fine-grained typography generation, suboptimal visual aesthetics and fidelity, and limited image resolutions. Specifically, the advancements of Seedream 3.0 stem from improvements across the entire pipeline, from data construction to model deployment. At the data stratum, we double the dataset using a defect-aware training paradigm and a dual-axis collaborative data-sampling framework. Furthermore, we adopt several effective techniques such as mixed-resolution training, cross-modality RoPE, representation alignment loss, and resolution-aware timestep sampling in the pre-training phase. During the post-training stage, we utilize diversified aesthetic captions in SFT, and a VLM-based reward model with scaling, thereby achieving outputs that well align with human preferences. Furthermore, Seedream 3.0 pioneers a novel acceleration paradigm. By employing consistent noise expectation and importance-aware timestep sampling, we achieve a 4 to 8 times speedup while maintaining image quality. Seedream 3.0 demonstrates significant improvements over Seedream 2.0: it enhances overall capabilities, in particular for text-rendering in complicated Chinese characters which is important to professional typography generation. In addition, it provides native high-resolution output (up to 2K), allowing it to generate images with high visual quality.

cs.CV↗

arXiv2025

Re-Evaluating Code LLM Benchmarks Under Semantic Mutation

In the era of large language models (LLMs), code benchmarks have become an important research area in software engineering and are widely used by practitioners. These benchmarks evaluate the performance of LLMs on specific code-related tasks, such as code understanding and generation. A critical step in constructing code benchmarks is the design of prompts. However, as existing code benchmarks typically rely on a single prompt template per task, they are prone to the issue of prompt sensitivity, where minor prompt variations could result in substantial performance variations, leading to unreliable evaluations of model capabilities. While previous studies have explored prompt sensitivity, their experimental designs and findings are limited to traditional natural language processing (NLP) tasks. In this paper, we present an empirical study to investigate prompt sensitivity in code benchmarks. We first propose a general framework that modifies prompt templates in a manner that preserves both their semantics and their structure as much as possible. Based on the framework, we conduct extensive experiments across eight code benchmark tasks on 10 representative open-source LLMs, with each task featuring 100 semantically similar prompt templates. We then analyze the evaluation results using various statistical metrics, focusing on both absolute and relative model performance. Our findings suggest that even slight prompt variations can lead to significant shifts in performance. Additionally, we observe that such variations can introduce inconsistencies in the performance rankings across different models. These insights highlight the need for considering prompt sensitivity when designing future code benchmarks, to ensure more reliable and accurate evaluation of LLM capabilities.

cs.SE↗

arXiv2025

CooPre: Cooperative Pretraining for V2X Cooperative Perception

Existing Vehicle-to-Everything (V2X) cooperative perception methods rely on accurate multi-agent 3D annotations. Nevertheless, it is time-consuming and expensive to collect and annotate real-world data, especially for V2X systems. In this paper, we present a self-supervised learning framwork for V2X cooperative perception, which utilizes the vast amount of unlabeled 3D V2X data to enhance the perception performance. Specifically, multi-agent sensing information is aggregated to form a holistic view and a novel proxy task is formulated to reconstruct the LiDAR point clouds across multiple connected agents to better reason multi-agent spatial correlations. Besides, we develop a V2X bird-eye-view (BEV) guided masking strategy which effectively allows the model to pay attention to 3D features across heterogeneous V2X agents (i.e., vehicles and infrastructure) in the BEV space. Noticeably, such a masking strategy effectively pretrains the 3D encoder with a multi-agent LiDAR point cloud reconstruction objective and is compatible with mainstream cooperative perception backbones. Our approach, validated through extensive experiments on representative datasets (i.e., V2X-Real, V2V4Real, and OPV2V) and multiple state-of-the-art cooperative perception methods (i.e., AttFuse, F-Cooper, and V2X-ViT), leads to a performance boost across all V2X settings. Notably, CooPre achieves a 4% mAP improvement on V2X-Real dataset and surpasses baseline performance using only 50% of the training data, highlighting its data efficiency. Additionally, we demonstrate the framework's powerful performance in cross-domain transferability and robustness under challenging scenarios. The code will be made publicly available at https://github.com/ucla-mobility/CooPre.

cs.CV↗

arXiv2025

Towards Understanding Bugs in Distributed Training and Inference Frameworks for Large Language Models

With the rapid development of large language models (LLMs), distributed training and inference frameworks like DeepSpeed have become essential for scaling model training and inference across multiple GPUs or nodes. However, the increasing complexity of these frameworks brings non-trivial software bugs, which may degrade training performance, cause unexpected failures, and result in significant resource waste. Understanding framework bugs' characteristics is fundamental for quality assurance, allowing the design of more effective debugging and repair methods. Thus, our paper conducts the first large-scale empirical analysis of 308 fixed bugs across three popular distributed training/inference frameworks: DeepSpeed, Megatron-LM, and Colossal-AI. We examine bug symptoms, root causes, bug identification and fixing efforts, and common low-effort fixing strategies. Additionally, the distributed nature of these frameworks introduces unique bug root causes, such as allocation strategy error and distributed communication error. Diagnosing and fixing complex bugs remains challenging due to factors like the disconnect between symptoms and root causes, high bug reproduction costs, and low-level or cross-component interactions. Interestingly, we observe that 48% of bug fixes require minimal code changes (<=10 LOC) and follow simple strategies such as conditional logic optimization, parameter handling enhancement, or version compatibility handling, indicating potential for automation. Based on these insights, we offer several implications for improving the reliability of both distributed training and inference frameworks and their dependent LLM projects, while also identifying opportunities to leverage LLM-based tools for automated debugging and repair.

cs.SE↗

arXiv2025

One-to-one correspondence reconstruction at the electron-positron Higgs factory

We propose one-to-one correspondence reconstruction for electron-positron Higgs factories. For each visible particle, one-to-one correspondence aims to associate relevant detector hits with only one reconstructed particle and accurately identify its species. To achieve this goal, we develop a novel detector concept featuring 5-dimensional calorimetry that provides spatial, energy, and time measurements for each hit, and a reconstruction framework that combines state-of-the-art particle flow and artificial intelligence algorithms. In the benchmark process of Higgs to di-jets, over 90% of visible energy can be successfully mapped into well-reconstructed particles that not only maintain a one-to-one correspondence relationship but also associate with the correct combination of cluster and track, improving the invariant mass resolution of hadronically decayed Higgs bosons by 25%. Performing simultaneous identification on these well-reconstructed particles, we observe efficiencies of 97% to nearly 100% for charged particles ($e^{\pm}$, $μ^{\pm}$, $π^{\pm}$, $K^{\pm}$, $p/\bar{p}$) and photons ($γ$), and 75% to 80% for neutral hadrons ($K_L^0$, $n$, $\bar{n}$). For physics measurements of Higgs to invisible and exotic decays, golden channels to probe new physics, one-to-one correspondence could enhance discovery power by 10% to up to a factor of two. This study demonstrates the necessity and feasibility of one-to-one correspondence reconstruction at electron-positron Higgs factories.

hep-ex↗

arXiv2025

SeedEdit 3.0: Fast and High-Quality Generative Image Editing

We introduce SeedEdit 3.0, in companion with our T2I model Seedream 3.0, which significantly improves over our previous SeedEdit versions in both aspects of edit instruction following and image content (e.g., ID/IP) preservation on real image inputs. Additional to model upgrading with T2I, in this report, we present several key improvements. First, we develop an enhanced data curation pipeline with a meta-info paradigm and meta-info embedding strategy that help mix images from multiple data sources. This allows us to scale editing data effectively, and meta information is helpfult to connect VLM with diffusion model more closely. Second, we introduce a joint learning pipeline for computing a diffusion loss and reward losses. Finally, we evaluate SeedEdit 3.0 on our testing benchmarks, for real/synthetic image editing, where it achieves a best trade-off between multiple aspects, yielding a high usability rate of 56.1%, compared to SeedEdit 1.6 (38.4%), GPT4o (37.1%) and Gemini 2.0 (30.3%).

cs.CV↗

arXiv2025

ActRef: Enhancing the Understanding of Python Code Refactoring with Action-Based Analysis

Refactoring, the process of improving the code structure of a software system without altering its behavior, is crucial for managing code evolution in software development. Identifying refactoring actions in source code is essential for understanding software evolution and guiding developers in maintaining and improving the code quality. This study presents an action-based Refactoring Analysis Framework named ActRef, a novel algorithm designed to advance the detection and understanding of Python refactorings through a unique code change action-based analysis of code changes. ActRef mining multiple refactoring types (e.g., move, rename, extract, and inline operations) based on diff actions, covering multiple granularity levels including variable, method, class, and module levels. By focusing on the code change actions, ActRef provides a Python-adaptive solution to detect intricate refactoring patterns. Our evaluation, conducted on 1,914 manually validated refactoring instances from 136 open-source Python projects. The evaluation results show that ActRef achieves high precision(0.80) and recall(0.92), effectively identifying multiple refactoring types. Compared with leading baselines, including PyRef, PyRef with MLRefScanner, DeepSeek-R1 and ChatGPT-4, ActRef consistently demonstrates superior performance in detecting Python refactorings across various types. While matching PyRef in runtime efficiency, ActRef supports a broader spectrum of refactoring types and more refactoring mining levels. ActRef shows an effective and scalable approach for mining refactorings in dynamic Python codebases and introduces a new perspective on understanding code.

cs.SE↗

arXiv2025

Seaweed-7B: Cost-Effective Training of Video Generation Foundation Model

This technical report presents a cost-efficient strategy for training a video generation foundation model. We present a mid-sized research model with approximately 7 billion parameters (7B) called Seaweed-7B trained from scratch using 665,000 H100 GPU hours. Despite being trained with moderate computational resources, Seaweed-7B demonstrates highly competitive performance compared to contemporary video generation models of much larger size. Design choices are especially crucial in a resource-constrained setting. This technical report highlights the key design decisions that enhance the performance of the medium-sized diffusion model. Empirically, we make two observations: (1) Seaweed-7B achieves performance comparable to, or even surpasses, larger models trained on substantially greater GPU resources, and (2) our model, which exhibits strong generalization ability, can be effectively adapted across a wide range of downstream applications either by lightweight fine-tuning or continue training. See the project page at https://seaweed.video/

cs.CV↗

arXiv2025

When Deep Learning Meets Information Retrieval-based Bug Localization: A Survey

Bug localization is a crucial aspect of software maintenance, running through the entire software lifecycle. Information retrieval-based bug localization (IRBL) identifies buggy code based on bug reports, expediting the bug resolution process for developers. Recent years have witnessed significant achievements in IRBL, propelled by the widespread adoption of deep learning (DL). To provide a comprehensive overview of the current state of the art and delve into key issues, we conduct a survey encompassing 61 IRBL studies leveraging DL. We summarize best practices in each phase of the IRBL workflow, undertake a meta-analysis of prior studies, and suggest future research directions. This exploration aims to guide further advancements in the field, fostering a deeper understanding and refining practices for effective bug localization. Our study suggests that the integration of DL in IRBL enhances the model's capacity to extract semantic and syntactic information from both bug reports and source code, addressing issues such as lexical gaps, neglect of code structure information, and cold-start problems. Future research avenues for IRBL encompass exploring diversity in programming languages, adopting fine-grained granularity, and focusing on real-world applications. Most importantly, although some studies have started using large language models for IRBL, there is still a need for more in-depth exploration and thorough investigation in this area.

cs.SE↗

arXiv2025

Understanding Practitioners' Expectations on Clear Code Review Comments

The code review comment (CRC) is pivotal in the process of modern code review. It provides reviewers with the opportunity to identify potential bugs, offer constructive feedback, and suggest improvements. Clear and concise code review comments (CRCs) facilitate the communication between developers and are crucial to the correct understanding of the identified issues and proposed solutions. Despite the importance of CRCs' clarity, there is still a lack of guidelines on what constitutes a good clarity and how to evaluate it. In this paper, we conduct a comprehensive study on understanding and evaluating the clarity of CRCs. We first derive a set of attributes related to the clarity of CRCs, namely RIE attributes (i.e., Relevance, Informativeness, and Expression), as well as their corresponding evaluation criteria based on our literature review and survey with practitioners. We then investigate the clarity of CRCs in open-source projects written in nine programming languages and find that a large portion (i.e., 28.8%) of the CRCs lack the clarity in at least one of the attributes. Finally, we explore the potential of automatically evaluating the clarity of CRCs by proposing ClearCRC. Experimental results show that ClearCRC with pre-trained language models is promising for effective evaluation of the clarity of CRCs, achieving a balanced accuracy up to 73.04% and a F-1 score up to 94.61%.

cs.SE↗

arXiv2025

Can LLMs Replace Human Evaluators? An Empirical Study of LLM-as-a-Judge in Software Engineering

Recently, large language models (LLMs) have been deployed to tackle various software engineering (SE) tasks like code generation, significantly advancing the automation of SE tasks. However, assessing the quality of these LLM-generated code and text remains challenging. The commonly used Pass@k metric necessitates extensive unit tests and configured environments, demands a high labor cost, and is not suitable for evaluating LLM-generated text. Conventional metrics like BLEU, which measure only lexical rather than semantic similarity, have also come under scrutiny. In response, a new trend has emerged to employ LLMs for automated evaluation, known as LLM-as-a-judge. These LLM-as-a-judge methods are claimed to better mimic human assessment than conventional metrics without relying on high-quality reference answers. Nevertheless, their exact human alignment in SE tasks remains unexplored. In this paper, we empirically explore LLM-as-a-judge methods for evaluating SE tasks, focusing on their alignment with human judgments. We select seven LLM-as-a-judge methods that utilize general-purpose LLMs, alongside two LLMs specifically fine-tuned for evaluation. After generating and manually scoring LLM responses on three recent SE datasets of code translation, code generation, and code summarization, we then prompt these methods to evaluate each response. Finally, we compare the scores generated by these methods with human evaluation. The results indicate that output-based methods reach the highest Pearson correlation of 81.32 and 68.51 with human scores in code translation and generation, achieving near-human evaluation, noticeably outperforming ChrF++, one of the best conventional metrics, at 34.23 and 64.92. Such output-based methods prompt LLMs to output judgments directly, and exhibit more balanced score distributions that resemble human score patterns. Finally, we provide...

cs.SE↗

arXiv2025

InSPE: Rapid Evaluation of Heterogeneous Multi-Modal Infrastructure Sensor Placement

Infrastructure sensing is vital for traffic monitoring at safety hotspots (e.g., intersections) and serves as the backbone of cooperative perception in autonomous driving. While vehicle sensing has been extensively studied, infrastructure sensing has received little attention, especially given the unique challenges of diverse intersection geometries, complex occlusions, varying traffic conditions, and ambient environments like lighting and weather. To address these issues and ensure cost-effective sensor placement, we propose Heterogeneous Multi-Modal Infrastructure Sensor Placement Evaluation (InSPE), a perception surrogate metric set that rapidly assesses perception effectiveness across diverse infrastructure and environmental scenarios with combinations of multi-modal sensors. InSPE systematically evaluates perception capabilities by integrating three carefully designed metrics, i.e., sensor coverage, perception occlusion, and information gain. To support large-scale evaluation, we develop a data generation tool within the CARLA simulator and also introduce Infra-Set, a dataset covering diverse intersection types and environmental conditions. Benchmarking experiments with state-of-the-art perception algorithms demonstrate that InSPE enables efficient and scalable sensor placement analysis, providing a robust solution for optimizing intelligent intersection infrastructure.

cs.RO↗

arXiv2025

Generating Mitigations for Downstream Projects to Neutralize Upstream Library Vulnerability

Third-party libraries are essential in software development as they prevent the need for developers to recreate existing functionalities. However, vulnerabilities within these libraries pose significant risks to dependent projects. Upgrading dependencies to secure versions is not feasible to neutralize vulnerabilities without patches or in projects with specific version requirements. Moreover, repairing the vulnerability proves challenging when the source code of the library is inaccessible. Both the state-of-the-art automatic vulnerability repair and automatic program repair methods fail to address this issue. Therefore, mitigating library vulnerabilities without source code and available patches is crucial for a swift response to potential security attacks. Existing tools encounter challenges concerning generalizability and functional security. In this study, we introduce LUMEN to mitigate library vulnerabilities in impacted projects. Upon disclosing a vulnerability, we retrieve existing workarounds to gather a resembling mitigation strategy. In cases where a resembling strategy is absent, we propose type-based strategies based on the vulnerability reproducing behavior and extract essential information from the vulnerability report to guide mitigation generation. Our assessment of LUMEN spans 121 impacted functions of 40 vulnerabilities, successfully mitigating 70.2% of the functions, which substantially outperforms our baseline in neutralizing vulnerabilities without functionality loss. Additionally, we conduct an ablation study to validate the rationale behind our resembling strategies and type-based strategies.

cs.SE↗

arXiv2025

Training-free Diffusion Acceleration with Bottleneck Sampling

Diffusion models have demonstrated remarkable capabilities in visual content generation but remain challenging to deploy due to their high computational cost during inference. This computational burden primarily arises from the quadratic complexity of self-attention with respect to image or video resolution. While existing acceleration methods often compromise output quality or necessitate costly retraining, we observe that most diffusion models are pre-trained at lower resolutions, presenting an opportunity to exploit these low-resolution priors for more efficient inference without degrading performance. In this work, we introduce Bottleneck Sampling, a training-free framework that leverages low-resolution priors to reduce computational overhead while preserving output fidelity. Bottleneck Sampling follows a high-low-high denoising workflow: it performs high-resolution denoising in the initial and final stages while operating at lower resolutions in intermediate steps. To mitigate aliasing and blurring artifacts, we further refine the resolution transition points and adaptively shift the denoising timesteps at each stage. We evaluate Bottleneck Sampling on both image and video generation tasks, where extensive experiments demonstrate that it accelerates inference by up to 3$\times$ for image generation and 2.5$\times$ for video generation, all while maintaining output quality comparable to the standard full-resolution sampling process across multiple evaluation metrics.

cs.CV↗