arXiv subjects

Yichen Li

Publications and source records attributed to Yichen Li.

At least 73 records · Page 4Linked to original sources

arXiv2025

Towards Robust Visual Continual Learning with Multi-Prototype Supervision

Language-guided supervision, which utilizes a frozen semantic target from a Pretrained Language Model (PLM), has emerged as a promising paradigm for visual Continual Learning (CL). However, relying on a single target introduces two critical limitations: 1) semantic ambiguity, where a polysemous category name results in conflicting visual representations, and 2) intra-class visual diversity, where a single prototype fails to capture the rich variety of visual appearances within a class. To this end, we propose MuproCL, a novel framework that replaces the single target with multiple, context-aware prototypes. Specifically, we employ a lightweight LLM agent to perform category disambiguation and visual-modal expansion to generate a robust set of semantic prototypes. A LogSumExp aggregation mechanism allows the vision model to adaptively align with the most relevant prototype for a given image. Extensive experiments across various CL baselines demonstrate that MuproCL consistently enhances performance and robustness, establishing a more effective path for language-guided continual learning.

cs.CV↗

arXiv2025

Trace Sampling 2.0: Code Knowledge Enhanced Span-level Sampling for Distributed Tracing

Distributed tracing is an essential diagnostic tool in microservice systems, but the sheer volume of traces places a significant burden on backend storage. A common approach to mitigating this issue is trace sampling, which selectively retains traces based on specific criteria, often preserving only anomalous ones. However, this method frequently discards valuable information, including normal traces that are essential for comparative analysis. To address this limitation, we introduce Trace Sampling 2.0, which operates at the span level while maintaining trace structure consistency. This approach allows for the retention of all traces while significantly reducing storage overhead. Based on this concept, we design and implement Autoscope, a span-level sampling method that leverages static analysis to extract execution logic, ensuring that critical spans are preserved without compromising structural integrity. We evaluated Autoscope on two open-source microservices. Our results show that it reduces trace size by 81.2% while maintaining 98.1% faulty span coverage, outperforming existing trace-level sampling methods. Furthermore, we demonstrate its effectiveness in root cause analysis, achieving an average improvement of 8.3%. These findings indicate that Autoscope can significantly enhance observability and storage efficiency in microservices, offering a robust solution for performance monitoring.

cs.SE↗

arXiv2025

Next Edit Prediction: Learning to Predict Code Edits from Context and Interaction History

The rapid advancement of large language models (LLMs) has led to the widespread adoption of AI-powered coding assistants integrated into a development environment. On one hand, low-latency code completion offers completion suggestions but is fundamentally constrained to the cursor's current position. On the other hand, chat-based editing can perform complex modifications, yet forces developers to stop their work, describe the intent in natural language, which causes a context-switch away from the code. This creates a suboptimal user experience, as neither paradigm proactively predicts the developer's next edit in a sequence of related edits. To bridge this gap and provide the seamless code edit suggestion, we introduce the task of Next Edit Prediction, a novel task designed to infer developer intent from recent interaction history to predict both the location and content of the subsequent edit. Specifically, we curate a high-quality supervised fine-tuning dataset and an evaluation benchmark for the Next Edit Prediction task. Then, we conduct supervised fine-tuning on a series of models and performed a comprehensive evaluation of both the fine-tuned models and other baseline models, yielding several novel findings. This work lays the foundation for a new interaction paradigm that proactively collaborate with developers by anticipating their following action, rather than merely reacting to explicit instructions. The code is available at https://github.com/lurf21/NextEditPrediction.

cs.SE↗

arXiv2025

Larger Is Not Always Better: Exploring Small Open-source Language Models in Logging Statement Generation

Developers use logging statements to create logs that document system behavior and aid in software maintenance. As such, high-quality logging is essential for effective maintenance; however, manual logging often leads to errors and inconsistency. Recent methods emphasize using large language models (LLMs) for automated logging statement generation, but these present privacy and resource issues, hindering their suitability for enterprise use. This paper presents the first large-scale empirical study evaluating small open-source language models (SOLMs) for automated logging statement generation. We evaluate four prominent SOLMs using various prompt strategies and parameter-efficient fine-tuning techniques, such as Low-Rank Adaptation (LoRA) and Retrieval-Augmented Generation (RAG). Our results show that fine-tuned SOLMs with LoRA and RAG prompts, particularly Qwen2.5-coder-14B, outperform existing tools and LLM baselines in predicting logging locations and generating high-quality statements, with robust generalization across diverse repositories. These findings highlight SOLMs as a privacy-preserving, efficient alternative for automated logging.

cs.SE↗

arXiv2025

DentalBench: Benchmarking and Advancing LLMs Capability for Bilingual Dentistry Understanding

Recent advances in large language models (LLMs) and medical LLMs (Med-LLMs) have demonstrated strong performance on general medical benchmarks. However, their capabilities in specialized medical fields, such as dentistry which require deeper domain-specific knowledge, remain underexplored due to the lack of targeted evaluation resources. In this paper, we introduce DentalBench, the first comprehensive bilingual benchmark designed to evaluate and advance LLMs in the dental domain. DentalBench consists of two main components: DentalQA, an English-Chinese question-answering (QA) benchmark with 36,597 questions spanning 4 tasks and 16 dental subfields; and DentalCorpus, a large-scale, high-quality corpus with 337.35 million tokens curated for dental domain adaptation, supporting both supervised fine-tuning (SFT) and retrieval-augmented generation (RAG). We evaluate 14 LLMs, covering proprietary, open-source, and medical-specific models, and reveal significant performance gaps across task types and languages. Further experiments with Qwen-2.5-3B demonstrate that domain adaptation substantially improves model performance, particularly on knowledge-intensive and terminology-focused tasks, and highlight the importance of domain-specific benchmarks for developing trustworthy and effective LLMs tailored to healthcare applications.

cs.CL↗

arXiv2025

Exploring Autonomous Agents: A Closer Look at Why They Fail When Completing Tasks

Autonomous agent systems powered by Large Language Models (LLMs) have demonstrated promising capabilities in automating complex tasks. However, current evaluations largely rely on success rates without systematically analyzing the interactions, communication mechanisms, and failure causes within these systems. To bridge this gap, we present a benchmark of 34 representative programmable tasks designed to rigorously assess autonomous agents. Using this benchmark, we evaluate three popular open-source agent frameworks combined with two LLM backbones, observing a task completion rate of approximately 50%. Through in-depth failure analysis, we develop a three-tier taxonomy of failure causes aligned with task phases, highlighting planning errors, task execution issues, and incorrect response generation. Based on these insights, we propose actionable improvements to enhance agent planning and self-diagnosis capabilities. Our failure taxonomy, together with mitigation advice, provides an empirical foundation for developing more robust and effective autonomous agent systems in the future.

cs.AI↗

arXiv2025

EPANet: Efficient Path Aggregation Network for Underwater Fish Detection

Underwater fish detection (UFD) remains a challenging task in computer vision due to low object resolution, significant background interference, and high visual similarity between targets and surroundings. Existing approaches primarily focus on local feature enhancement or incorporate complex attention mechanisms to highlight small objects, often at the cost of increased model complexity and reduced efficiency. To address these limitations, we propose an efficient path aggregation network (EPANet), which leverages complementary feature integration to achieve accurate and lightweight UFD. EPANet consists of two key components: an efficient path aggregation feature pyramid network (EPA-FPN) and a multi-scale diverse-division short path bottleneck (MS-DDSP bottleneck). The EPA-FPN introduces long-range skip connections across disparate scales to improve semantic-spatial complementarity, while cross-layer fusion paths are adopted to enhance feature integration efficiency. The MS-DDSP bottleneck extends the conventional bottleneck structure by introducing finer-grained feature division and diverse convolutional operations, thereby increasing local feature diversity and representation capacity. Extensive experiments on benchmark UFD datasets demonstrate that EPANet outperforms state-of-the-art methods in terms of detection accuracy and inference speed, while maintaining comparable or even lower parameter complexity.

cs.CV↗

arXiv2025

FairSteer: Inference Time Debiasing for LLMs with Dynamic Activation Steering

Large language models (LLMs) are prone to capturing biases from training corpus, leading to potential negative social impacts. Existing prompt-based debiasing methods exhibit instability due to their sensitivity to prompt changes, while fine-tuning-based techniques incur substantial computational overhead and catastrophic forgetting. In this paper, we propose FairSteer, a novel inference-time debiasing framework without requiring customized prompt design or model retraining. Motivated by the linear representation hypothesis, our preliminary investigation demonstrates that fairness-related features can be encoded into separable directions in the hidden activation space. FairSteer operates in three steps: biased activation detection, debiasing steering vector (DSV) computation, and dynamic activation steering. Specifically, it first trains a lightweight linear classifier to detect bias signatures in activations, and then computes DSVs as intervention directions derived from small contrastive prompt pairs. Subsequently, it performs debiasing by adjusting activations with DSVs in the inference stage. Comprehensive evaluation with six LLMs demonstrates the superiority of FairSteer across question-answering, counterfactual input evaluation and open-ended text generation tasks. Code will be released.

cs.CL↗

arXiv2025

CCISolver: End-to-End Detection and Repair of Method-Level Code-Comment Inconsistency

Comments within code serve as a crucial foundation for software documentation, facilitating developers to communicate and understand the code effectively. However, code-comment inconsistency (CCI) can negatively affect software development, testing, and maintenance. Recent efforts to mitigate this issue have emerged, but existing studies often suffer from inaccurate datasets and inadequate solutions, weakening their practical effectiveness. In this study, we first conduct a quantitative analysis of existing datasets, revealing a substantial portion of sampled data are mislabeled. To address these data limitations, we introduce CCIBench, a refined dataset comprising high-quality data, to support the training and evaluation of method-level CCI methods. Furthermore, we present an innovative end-to-end LLM-based framework, CCISolver, designed to improve code quality by identifying and rectifying CCIs. Comprehensive evaluations demonstrate CCISolver's superior performance. For detection, it establishes a new state-of-the-art with an F1-score of 89.54%. In fixing task, it achieves a remarkable 18.84% relative improvement in GLEU score over the strongest baseline. This superiority is confirmed by human evaluation, where CCISolver's fixing success rate of 0.6533 significantly surpasses existing methods. Critically, in a practical end-to-end setting, CCISolver's innovative architecture is approximately 36% faster for inference than the baseline model, underscoring its scalability and real-world applicability.

cs.SE↗

arXiv2025

MAFE R-CNN: Selecting More Samples to Learn Category-aware Features for Small Object Detection

Small object detection in intricate environments has consistently represented a major challenge in the field of object detection. In this paper, we identify that this difficulty stems from the detectors' inability to effectively learn discriminative features for objects of small size, compounded by the complexity of selecting high-quality small object samples during training, which motivates the proposal of the Multi-Clue Assignment and Feature Enhancement R-CNN.Specifically, MAFE R-CNN integrates two pivotal components.The first is the Multi-Clue Sample Selection (MCSS) strategy, in which the Intersection over Union (IoU) distance, predicted category confidence, and ground truth region sizes are leveraged as informative clues in the sample selection process. This methodology facilitates the selection of diverse positive samples and ensures a balanced distribution of object sizes during training, thereby promoting effective model learning.The second is the Category-aware Feature Enhancement Mechanism (CFEM), where we propose a simple yet effective category-aware memory module to explore the relationships among object features. Subsequently, we enhance the object feature representation by facilitating the interaction between category-aware features and candidate box features.Comprehensive experiments conducted on the large-scale small object dataset SODA validate the effectiveness of the proposed method. The code will be made publicly available.

cs.CV↗

arXiv2025

Rehearsal-Free Continual Federated Learning with Synergistic Synaptic Intelligence

Continual Federated Learning (CFL) allows distributed devices to collaboratively learn novel concepts from continuously shifting training data while avoiding knowledge forgetting of previously seen tasks. To tackle this challenge, most current CFL approaches rely on extensive rehearsal of previous data. Despite effectiveness, rehearsal comes at a cost to memory, and it may also violate data privacy. Considering these, we seek to apply regularization techniques to CFL by considering their cost-efficient properties that do not require sample caching or rehearsal. Specifically, we first apply traditional regularization techniques to CFL and observe that existing regularization techniques, especially synaptic intelligence, can achieve promising results under homogeneous data distribution but fail when the data is heterogeneous. Based on this observation, we propose a simple yet effective regularization algorithm for CFL named FedSSI, which tailors the synaptic intelligence for the CFL with heterogeneous data settings. FedSSI can not only reduce computational overhead without rehearsal but also address the data heterogeneity issue. Extensive experiments show that FedSSI achieves superior performance compared to state-of-the-art methods.

cs.LG↗

arXiv2025

FedRE: Robust and Effective Federated Learning with Privacy Preference

Despite Federated Learning (FL) employing gradient aggregation at the server for distributed training to prevent the privacy leakage of raw data, private information can still be divulged through the analysis of uploaded gradients from clients. Substantial efforts have been made to integrate local differential privacy (LDP) into the system to achieve a strict privacy guarantee. However, existing methods fail to take practical issues into account by merely perturbing each sample with the same mechanism while each client may have their own privacy preferences on privacy-sensitive information (PSI), which is not uniformly distributed across the raw data. In such a case, excessive privacy protection from private-insensitive information can additionally introduce unnecessary noise, which may degrade the model performance. In this work, we study the PSI within data and develop FedRE, that can simultaneously achieve robustness and effectiveness benefits with LDP protection. More specifically, we first define PSI with regard to the privacy preferences of each client. Then, we optimize the LDP by allocating less privacy budget to gradients with higher PSI in a layer-wise manner, thus providing a stricter privacy guarantee for PSI. Furthermore, to mitigate the performance degradation caused by LDP, we design a parameter aggregation mechanism based on the distribution of the perturbed information. We conducted experiments with text tamper detection on T-SROIE and DocTamper datasets, and FedRE achieves competitive performance compared to state-of-the-art methods.

cs.LG↗

arXiv2025

Unleashing the Power of Continual Learning on Non-Centralized Devices: A Survey

Non-Centralized Continual Learning (NCCL) has become an emerging paradigm for enabling distributed devices such as vehicles and servers to handle streaming data from a joint non-stationary environment. To achieve high reliability and scalability in deploying this paradigm in distributed systems, it is essential to conquer challenges stemming from both spatial and temporal dimensions, manifesting as distribution shifts, catastrophic forgetting, heterogeneity, and privacy issues. This survey focuses on a comprehensive examination of the development of the non-centralized continual learning algorithms and the real-world deployment across distributed devices. We begin with an introduction to the background and fundamentals of non-centralized learning and continual learning. Then, we review existing solutions from three levels to represent how existing techniques alleviate the catastrophic forgetting and distribution shift. Additionally, we delve into the various types of heterogeneity issues, security, and privacy attributes, as well as real-world applications across three prevalent scenarios. Furthermore, we establish a large-scale benchmark to revisit this problem and analyze the performance of the state-of-the-art NCCL approaches. Finally, we discuss the important challenges and future research directions in NCCL.

cs.LG↗

arXiv2025

LLMPrism: Black-box Performance Diagnosis for Production LLM Training Platforms

Large Language Models (LLMs) have brought about revolutionary changes in diverse fields, rendering LLM training of utmost importance for modern enterprises. To meet this demand, multi-tenant large-scale LLM training platforms have been built to offer LLM training services. Nevertheless, due to the complexity and synchronous nature of LLM training process, performance issues occur frequently and can result in substantial resource wastage. The limited visibility from the perspective of platform providers impedes existing profiling methods and poses challenges to the monitoring and diagnosis of the performance of LLM training jobs. For the first time, this paper proposes the utilization of underlying network flow data to reconstruct the training timelines of jobs based on the distinct characteristics in the LLM training procedure. We design LLMPrism, the first black-box performance diagnosis system for LLM training platforms. By progressively recognizing LLM training jobs, identifying their parallelism strategies, and reconstructing the training timelines, LLMPrism achieves non-intrusive, lightweight, and continuous monitoring of LLM training systems. Leveraging this monitoring capability, it further effectively diagnoses potential performance issues. Since Oct. 2024, LLMPrism has been deployed on our large-scale production Platform-X, in which the evaluations and deployment experiences demonstrate that LLMPrism can achieve accurate timeline reconstruction with an error within 0.3% and effectively diagnose various performance issues.

cs.SE↗

arXiv2025

LogUpdater: Automated Detection and Repair of Specific Defects in Logging Statements

Developers use logging statements to track software runtime behaviors and system status. Yet, unclear or misleading logs can hide true execution patterns and hinder software maintenance. Current research on logging statement issues is limited, often only spotting one defect type and relying on manual corrections instead of automation. To bridge this gap, we conduct a study to identify four logging statement defect types by analyzing log-centric changes. Then we introduce LogUpdater, a two-stage framework for automatically detecting and updating these log defects. In the offline phase, LogUpdater builds a classifier using synthetic defective logs to spot defect types. During online testing, this classifier assesses if and how logs in code snippets need improvement. LogUpdater then uses type-aware prompts from past logging updates to suggest fixes via a recommendation framework based on LLMs. Results show strong defect detection with an F1 score of 0.625. It also greatly improves static text and dynamic variable suggestions by 48.12% and 24.90%, respectively. LogUpdater successfully recommends updates 61.49% of the time on new projects. We reported 40 problematic logs and their fixes on GitHub, leading to 25 merged changes across 11 projects.

cs.SE↗

arXiv2025

Performance of plastic scintillator modules for top veto tracker at Taishan Antineutrino Observatory

The Taishan Antineutrino Observatory (TAO) experiment incorporates a top veto tracker (TVT) system comprising 160 modules, each composed of plastic scintillator (PS) strips, embedded wavelength shifting fibers (WLS-fibers), and silicon photomultipliers (SiPMs). This article highlights the performance of all produced modules following the production and readout/trigger design, providing insights for scintillation detectors with WLS-fibers. Three kinds of trigger modes and its efficiency have been defined to comprehensively evaluate the performance of this unique design, which has been verified for the batch production, along with comprehensive measurement strategies and quality inspection methods. In "module" mode, the detection(tagging) efficiency of the PS exceeds 99.67\% at a 30 photoelectron threshold, and even in "AND" mode, it surpasses 99.60\% at a 15 photoelectron threshold. The muon tagging efficiency meets TAO's requirements. The production and performance of the PS module set a benchmark for other experiments, with optimized optical fiber arrangements that enhance light yield and muon detection efficiency.

physics.ins-det↗

arXiv2025

COCA: Generative Root Cause Analysis for Distributed Systems with Code Knowledge

Runtime failures are commonplace in modern distributed systems. When such issues arise, users often turn to platforms such as Github or JIRA to report them and request assistance. Automatically identifying the root cause of these failures is critical for ensuring high reliability and availability. However, prevailing automatic root cause analysis (RCA) approaches rely significantly on comprehensive runtime monitoring data, which is often not fully available in issue platforms. Recent methods leverage large language models (LLMs) to analyze issue reports, but their effectiveness is limited by incomplete or ambiguous user-provided information. To obtain more accurate and comprehensive RCA results, the core idea of this work is to extract additional diagnostic clues from code to supplement data-limited issue reports. Specifically, we propose COCA, a code knowledge enhanced root cause analysis approach for issue reports. Based on the data within issue reports, COCA intelligently extracts relevant code snippets and reconstructs execution paths, providing a comprehensive execution context for further RCA. Subsequently, COCA constructs a prompt combining historical issue reports along with profiled code knowledge, enabling the LLMs to generate detailed root cause summaries and localize responsible components. Our evaluation on datasets from five real-world distributed systems demonstrates that COCA significantly outperforms existing methods, achieving a 28.3% improvement in root cause localization and a 22.0% improvement in root cause summarization. Furthermore, COCA's performance consistency across various LLMs underscores its robust generalizability.

cs.SE↗

arXiv2025

L4: Diagnosing Large-scale LLM Training Failures via Automated Log Analysis

As Large Language Models (LLMs) show their capabilities across various applications, training customized LLMs has become essential for modern enterprises. However, due to the complexity of LLM training, which requires massive computational resources and extensive training time, failures are inevitable during the training process. These failures result in considerable waste of resource and time, highlighting the critical need for effective and efficient failure diagnosis to reduce the cost of LLM training. In this paper, we present the first empirical study on the failure reports of 428 LLM training failures in our production Platform-X between May 2023 and April 2024. Our study reveals that hardware and user faults are the predominant root causes, and current diagnosis processes rely heavily on training logs. Unfortunately, existing log-based diagnostic methods fall short in handling LLM training logs. Considering the unique features of LLM training, we identify three distinct patterns of LLM training logs: cross-job, spatial, and temporal patterns. We then introduce our Log-based Large-scale LLM training failure diagnosis framework, L4, which can automatically extract failure-indicating information (i.e., log events, nodes, stages, and iterations) from extensive training logs, thereby reducing manual effort and facilitating failure recovery. Experimental results on real-world datasets show that L4 outperforms existing approaches in identifying failure-indicating logs and localizing faulty nodes. Furthermore, L4 has been applied in Platform-X and demonstrated its effectiveness in enabling accurate and efficient failure diagnosis.

cs.SE↗