arXiv subjects

Xuan Xie

Publications and source records attributed to Xuan Xie.

At least 19 recordsLinked to original sources

arXiv2026

Understanding Software Defect Prediction: A Large-scale Empirical Study Across Uncertainty Quantification and Performance Evaluation

Software defect prediction (SDP) classifiers produce probabilities used for inspection prioritization, threshold tuning, and risk communication. Probability-based uncertainty quantification (UQ) characterizes prediction confidence, but whether common UQ metrics reliably indicate performance and calibration remains unclear. We conducted a large-scale empirical study of probability-based UQ for SDP. We evaluated five UQ metrics, six performance metrics, and three calibration metrics for 16 representative classifiers. We analyzed these relationships under two prediction settings: within-project defect prediction (WPDP), using 36 benchmark datasets, and cross-project defect prediction (CPDP), using 32 feature-compatible datasets. Results showed that UQ was highly context-dependent. Under WPDP, UQ correlated more consistently with false positive rate and AUC than with MCC, F1 score, and other metrics; these correlations also varied across classifier categories and dataset collections. Performance and calibration were related but not interchangeable; classifiers with strong discrimination could still exhibit large calibration error. Under CPDP, several UQ-performance and UQ-calibration correlations weakened or reversed, indicating that uncertainty signals do not reliably transfer across projects. Thus, UQ should be evaluated against specific performance objectives. Calibration should be assessed independently using multiple metrics. Transferred probabilities should be revalidated before guiding quality-assurance decisions.

cs.SE↗

arXiv2026

MASTOR: A Multi-Agent Approach to Semantic Test Oracle Generation for RESTful APIs

Existing automated RESTful API testing approaches commonly rely on simple checks (e.g., HTTP status codes, schema conformance), which are insufficient for detecting semantic faults, business logic violations, and state-dependent inconsistencies. To address this, we propose MASTOR, a Multi-Agent approach for generating Semantic Test Oracles for RESTful APIs based on implementation source code. MASTOR consists of two phases: source analysis and oracle generation. The former employs a source extraction agent to construct a source context for each endpoint operation by analyzing a transitive import closure of relevant source files. The latter employs two parallel oracle-generation paths over the collected contexts: a single-operation path producing status and field oracles per operation, and a multi-operation path generating behavioral consistency oracles for operation sequences by leveraging cross-operation semantic associations. Both paths apply a challenger-agent review, where a dedicated reviewer identifies weaknesses and issues improvement hints to guide targeted regeneration, followed by oracle normalization to filter out structurally invalid oracles. We evaluated MASTOR on a benchmark of 13 open-source RESTful API projects (296 operations, 251,303 lines of code) from the WFD and PRAB datasets. MASTOR achieved an average mutation score of 75.4%, generating 10,022 oracles. These oracles were translated into executable assertions via ToJUnit and ToPostmanAssertify, and into human-readable descriptions via ToReadable. In a baseline comparison on 50 selected operations, MASTOR outperformed Direct Prompting by 30.1 percentage points (69.9% vs. 39.8%) and SATORI by 49.4 percentage points (69.9% vs. 20.5%).

cs.SE↗

arXiv2026

GMENet: Generative Mixture of Experts Network for Multi-Center Glioma Diagnosis with Incomplete Imaging Sequences

Contemporary glioma diagnosis integrates molecular features with histopathology to guide clinical decision-making. However, in clinical settings, divergent imaging protocols result in incomplete MRI sequences, leading to two primary challenges: forcing existing frameworks to discard a large portion of clinical data during training and consequently limiting their clinical applicability. To address these limitations, we propose GMENet, a Generative Mixture of Experts Network for multi-center glioma diagnosis with incomplete imaging sequences. Firstly, we design a Cross-attention-based Gated Generation Module that synthesizes missing sequence features from available sequences via cross-attention and dynamic gating mechanisms, incorporating a cycle-consistency loss to preserve semantic integrity. Secondly, we introduce a Dynamically Weighted Experts Fusion Module that performs mixture-of-experts interaction and confidence-aware fusion over original and synthesized dual-sequence features for multi-task prediction. We evaluate GMENet on a multi-center cohort of 1,241 subjects from four in-house datasets and two public repositories. Experiments show that GMENet expands clinically usable training data by 97\%, relative to complete-sequence-only data. Furthermore, it consistently outperforms state-of-the-art methods trained on complete data, demonstrating improved robustness under cross-center distribution shifts.

eess.IV↗

arXiv2026

KAT-Coder-V2 Technical Report

We present KAT-Coder-V2, an agentic coding model developed by the KwaiKAT team at Kuaishou. KAT-Coder-V2 adopts a "Specialize-then-Unify" paradigm that decomposes agentic coding into five expert domains - SWE, WebCoding, Terminal, WebSearch, and General - each undergoing independent supervised fine-tuning and reinforcement learning, before being consolidated into a single model via on-policy distillation. We develop KwaiEnv, a modular infrastructure sustaining tens of thousands of concurrent sandbox instances, and scale RL training along task complexity, intent alignment, and scaffold generalization. We further propose MCLA for stabilizing MoE RL training and Tree Training for eliminating redundant computation over tree-structured trajectories with up to 6.2x speedup. KAT-Coder-V2 achieves 79.6% on SWE-bench Verified (vs. Claude Opus 4.6 at 80.8%), 88.7 on PinchBench (surpassing GLM-5 and MiniMax M2.7), ranks first across all three frontend aesthetics scenarios, and maintains strong generalist scores on Terminal-Bench Hard (46.8) and tau^2-Bench (93.9). Our model is publicly available at https://streamlake.com/product/kat-coder.

cs.CL↗

arXiv2026

Prompt Group-Aware Training for Robust Text-Guided Nuclei Segmentation

Foundation models such as Segment Anything Model 3 (SAM3) enable flexible text-guided medical image segmentation, yet their predictions remain highly sensitive to prompt formulation. Even semantically equivalent descriptions can yield inconsistent masks, limiting reliability in clinical and pathology workflows. We reformulate prompt sensitivity as a group-wise consistency problem. Semantically related prompts are organized into \emph{prompt groups} sharing the same ground-truth mask, and a prompt group-aware training framework is introduced for robust text-guided nuclei segmentation. The approach combines (i) a quality-guided group regularization that leverages segmentation loss as an implicit ranking signal, and (ii) a logit-level consistency constraint with a stop-gradient strategy to align predictions within each group. The method requires no architectural modification and leaves inference unchanged. Extensive experiments on multi-dataset nuclei benchmarks show consistent gains under textual prompting and markedly reduced performance variance across prompt quality levels. On six zero-shot cross-dataset tasks, our method improves Dice by an average of 2.16 points. These results demonstrate improved robustness and generalization for vision-language segmentation in computational pathology.

cs.CV↗

arXiv2026

AISA: Awakening Intrinsic Safety Awareness in Large Language Models against Jailbreak Attacks

Large language models (LLMs) remain vulnerable to jailbreak prompts that elicit harmful or policy-violating outputs, while many existing defenses rely on expensive fine-tuning, intrusive prompt rewriting, or external guardrails that add latency and can degrade helpfulness. We present AISA, a lightweight, single-pass defense that activates safety behaviors already latent inside the model rather than treating safety as an add-on. AISA first localizes intrinsic safety awareness via spatiotemporal analysis and shows that intent-discriminative signals are broadly encoded, with especially strong separability appearing in the scaled dot-product outputs of specific attention heads near the final structural tokens before generation. Using a compact set of automatically selected heads, AISA extracts an interpretable prompt-risk score with minimal overhead, achieving detector-level performance competitive with strong proprietary baselines on small (7B) models. AISA then performs logits-level steering: it modulates the decoding distribution in proportion to the inferred risk, ranging from normal generation for benign prompts to calibrated refusal for high-risk requests -- without changing model parameters, adding auxiliary modules, or requiring multi-pass inference. Extensive experiments spanning 13 datasets, 12 LLMs, and 14 baselines demonstrate that AISA improves robustness and transfer while preserving utility and reducing false refusals, enabling safer deployment even for weakly aligned or intentionally risky model variants.

cs.CR↗

arXiv2025

DaGRPO: Rectifying Gradient Conflict in Reasoning via Distinctiveness-Aware Group Relative Policy Optimization

The evolution of Large Language Models (LLMs) has catalyzed a paradigm shift from superficial instruction following to rigorous long-horizon reasoning. While Group Relative Policy Optimization (GRPO) has emerged as a pivotal mechanism for eliciting such post-training reasoning capabilities due to its exceptional performance, it remains plagued by significant training instability and poor sample efficiency. We theoretically identify the root cause of these issues as the lack of distinctiveness within on-policy rollouts: for routine queries, highly homogeneous samples induce destructive gradient conflicts; whereas for hard queries, the scarcity of valid positive samples results in ineffective optimization. To bridge this gap, we propose Distinctiveness-aware Group Relative Policy Optimization (DaGRPO). DaGRPO incorporates two core mechanisms: (1) Sequence-level Gradient Rectification, which utilizes fine-grained scoring to dynamically mask sample pairs with low distinctiveness, thereby eradicating gradient conflicts at the source; and (2) Off-policy Data Augmentation, which introduces high-quality anchors to recover training signals for challenging tasks. Extensive experiments across 9 mathematical reasoning and out-of-distribution (OOD) generalization benchmarks demonstrate that DaGRPO significantly surpasses existing SFT, GRPO, and hybrid baselines, achieving new state-of-the-art performance (e.g., a +4.7% average accuracy gain on math benchmarks). Furthermore, in-depth analysis confirms that DaGRPO effectively mitigates gradient explosion and accelerates the emergence of long-chain reasoning capabilities.

cs.AI↗

arXiv2025

Outlier Smoothing with Closed-Form Rotations for W4A4 Large Language Model Quantization

Large Language Models (LLMs) quantization facilitates deploying LLMs in resource-limited settings, but existing methods that combine incompatible gradient optimization and quantization truncation lead to serious convergence pathology. This prolongs quantization time and degrades LLMs' task performance. Our studies confirm that Straight-Through Estimator (STE) on Stiefel manifolds introduce non-smoothness and gradient noise, obstructing optimization convergence and blocking high-fidelity quantized LLM development despite extensive training. To tackle the above limitations, we propose SingleQuant, a single-pass quantization framework that decouples from quantization truncation, thereby eliminating the above non-smoothness and gradient noise factors. Specifically, SingleQuant constructs Alignment Rotation Transformation (ART) and Uniformity Rotation Transformation (URT) targeting distinct activation outliers, where ART achieves smoothing of outlier values via closed-form optimal rotations, and URT reshapes distributions through geometric mapping. Both matrices comprise strictly formulated Givens rotations with predetermined dimensions and rotation angles, enabling promising LLMs task performance within a short time. Experimental results demonstrate SingleQuant's superiority over the selected baselines across diverse tasks on 7B-70B LLMs. To be more precise, SingleQuant enables quantized LLMs to achieve higher task performance while necessitating less time for quantization. For example, when quantizing LLaMA-2-13B, SingleQuant achieves 1,400$\times$ quantization speedup and increases +0.57\% average task performance compared to the selected best baseline.

cs.LG↗

arXiv2025

SAMPO-Path: Segmentation Intent-Aligned Preference Optimization for Pathology Foundation Model Segmentation

Foundation models have shown strong performance in multi-object segmentation with visual prompts, yet histopathology images remain challenging due to high cellular density, heterogeneity, and the gap between pixel-level supervision and clinical segmentation intent (e.g., selectively segmenting nuclei of a specific type). In practice, such intents are expressed through diverse and noisy prompts, causing prompt-intent misalignment and inconsistent predictions. We introduce SAMPO (Segmentation Anything Model with Preference Optimization), a preference-aligned fine-tuning framework that explicitly aligns pathology foundation models with clinical segmentation intent. SAMPO is the first to adapt Direct Preference Optimization (DPO) to pure vision foundation models, enabling accurate segmentation from minimal and imperfect prompts. The framework features three key components: (1) online prompt-centric preference mining to synthesize preference pairs across prompt qualities; (2) multi-mask preference learning to leverage output ambiguity for fine-grained ranking supervision; and (3) a hybrid loss combining preference optimization with pixel-level supervision for stable training. Trained on two datasets covering four tasks and evaluated on corresponding test sets and 12 external validation datasets, SAMPO consistently improves segmentation accuracy, robustness to prompt variations, and clinical intent adherence in dense histopathology images.

cs.CV↗

arXiv2025

MindShot: Multi-Shot Video Reconstruction from fMRI with LLM Decoding

Reconstructing dynamic videos from fMRI is important for understanding visual cognition and enabling vivid brain-computer interfaces. However, current methods are critically limited to single-shot clips, failing to address the multi-shot nature of real-world experiences. Multi-shot reconstruction faces fundamental challenges: fMRI signal mixing across shots, the temporal resolution mismatch between fMRI and video obscuring rapid scene changes, and the lack of dedicated multi-shot fMRI-video datasets. To overcome these limitations, we propose a novel divide-and-decode framework for multi-shot fMRI video reconstruction. Our core innovations are: (1) A shot boundary predictor module explicitly decomposing mixed fMRI signals into shot-specific segments. (2) Generative keyframe captioning using LLMs, which decodes robust textual descriptions from each segment, overcoming temporal blur by leveraging high-level semantics. (3) Novel large-scale data synthesis (20k samples) from existing datasets. Experimental results demonstrate our framework outperforms state-of-the-art methods in multi-shot reconstruction fidelity. Ablation studies confirm the critical role of fMRI decomposition and semantic captioning, with decomposition significantly improving decoded caption CLIP similarity by 71.8%. This work establishes a new paradigm for multi-shot fMRI reconstruction, enabling accurate recovery of complex visual narratives through explicit decomposition and semantic prompting.

cs.CV↗

arXiv2025

Chirally Frustrated Superradiant Phases in a Jaynes-Cummings Trimer

We investigate the emergence of frustrated quantum phases in a Jaynes-Cummings (JC) trimer with complex hopping amplitudes between the cavities, which represents the smallest frustrated unit in light-matter systems. The complex hopping amplitudes that can be engineered via synthetic gauge fields introduce chiral effects and geometric frustration into the system. We obtain analytic solutions in the semiclassical limit and map out the phase diagram of this model, featuring one normal and three distinct superradiant phases. Among these phases, a chirally frustrated superradiant phase emerges, characterized by broken chiral and translational symmetries and unidirectional photon flow. These results reveal how frustration and symmetry breaking can arise in JC systems with synthetic gauge fields and ultrastrong coupling.

quant-ph↗

arXiv2025

Agent Network Protocol Technical White Paper

With the development of large models and autonomous decision-making AI, agents are rapidly becoming the new entities of the internet, following mobile apps. However, existing internet infrastructure is primarily designed for human interaction, creating data silos, unfriendly interfaces, and high collaboration costs among agents, making it difficult to support the needs for large-scale agent interconnection and collaboration. The internet is undergoing a profound transformation, showing four core trends: agents replacing traditional software, universal agent interconnection, native protocol-based connections, and autonomous agent organization and collaboration. To align with these trends, Agent Network Protocol (ANP) proposes a new generation of communication protocols for the Agentic Web. ANP adheres to AI-native design, maintains compatibility with existing internet protocols, adopts a modular composable architecture, follows minimalist yet extensible principles, and enables rapid deployment based on existing infrastructure. Through a three-layer protocol system--identity and encrypted communication layer, meta-protocol negotiation layer, and application protocol layer--ANP. systematically solves the problems of agent identity authentication, dynamic negotiation, and capability discovery interoperability.

cs.NI↗

arXiv2025

EAT: QoS-Aware Edge-Collaborative AIGC Task Scheduling via Attention-Guided Diffusion Reinforcement Learning

The growth of Artificial Intelligence (AI) and large language models has enabled the use of Generative AI (GenAI) in cloud data centers for diverse AI-Generated Content (AIGC) tasks. Models like Stable Diffusion introduce unavoidable delays and substantial resource overhead, which are unsuitable for users at the network edge with high QoS demands. Deploying AIGC services on edge servers reduces transmission times but often leads to underutilized resources and fails to optimally balance inference latency and quality. To address these issues, this paper introduces a QoS-aware \underline{E}dge-collaborative \underline{A}IGC \underline{T}ask scheduling (EAT) algorithm. Specifically: 1) We segment AIGC tasks and schedule patches to various edge servers, formulating it as a gang scheduling problem that balances inference latency and quality while considering server heterogeneity, such as differing model distributions and cold start issues. 2) We propose a reinforcement learning-based EAT algorithm that uses an attention layer to extract load and task queue information from edge servers and employs a diffusion-based policy network for scheduling, efficiently enabling model reuse. 3) We develop an AIGC task scheduling system that uses our EAT algorithm to divide tasks and distribute them across multiple edge servers for processing. Experimental results based on our system and large-scale simulations show that our EAT algorithm can reduce inference latency by up to 56\% compared to baselines. We release our open-source code at https://github.com/zzf1955/EAT.

cs.DC↗

arXiv2025

Fine-grained Testing for Autonomous Driving Software: a Study on Autoware with LLM-driven Unit Testing

Testing autonomous driving systems (ADS) is critical to ensuring their reliability and safety. Existing ADS testing works focuses on designing scenarios to evaluate system-level behaviors, while fine-grained testing of ADS source code has received comparatively little attention. To address this gap, we present the first study on testing, specifically unit testing, for ADS source code. Our study focuses on an industrial ADS framework, Autoware. We analyze both human-written test cases and those generated by large language models (LLMs). Our findings reveal that human-written test cases in Autoware exhibit limited test coverage, and significant challenges remain in applying LLM-generated tests for Autoware unit testing. To overcome these challenges, we propose AwTest-LLM, a novel approach to enhance test coverage and improve test case pass rates across Autoware packages.

cs.SE↗

arXiv2024

Cross-Patient Pseudo Bags Generation and Curriculum Contrastive Learning for Imbalanced Multiclassification of Whole Slide Image

Pathology computing has dramatically improved pathologists' workflow and diagnostic decision-making processes. Although computer-aided diagnostic systems have shown considerable value in whole slide image (WSI) analysis, the problem of multi-classification under sample imbalance remains an intractable challenge. To address this, we propose learning fine-grained information by generating sub-bags with feature distributions similar to the original WSIs. Additionally, we utilize a pseudo-bag generation algorithm to further leverage the abundant and redundant information in WSIs, allowing efficient training in unbalanced-sample multi-classification tasks. Furthermore, we introduce an affinity-based sample selection and curriculum contrastive learning strategy to enhance the stability of model representation learning. Unlike previous approaches, our framework transitions from learning bag-level representations to understanding and exploiting the feature distribution of multi-instance bags. Our method demonstrates significant performance improvements on three datasets, including tumor classification and lymph node metastasis. On average, it achieves a 4.39-point improvement in F1 score compared to the second-best method across the three tasks, underscoring its superior performance.

cs.CV↗

arXiv2024

LeCov: Multi-level Testing Criteria for Large Language Models

Large Language Models (LLMs) are widely used in many different domains, but because of their limited interpretability, there are questions about how trustworthy they are in various perspectives, e.g., truthfulness and toxicity. Recent research has started developing testing methods for LLMs, aiming to uncover untrustworthy issues, i.e., defects, before deployment. However, systematic and formalized testing criteria are lacking, which hinders a comprehensive assessment of the extent and adequacy of testing exploration. To mitigate this threat, we propose a set of multi-level testing criteria, LeCov, for LLMs. The criteria consider three crucial LLM internal components, i.e., the attention mechanism, feed-forward neurons, and uncertainty, and contain nine types of testing criteria in total. We apply the criteria in two scenarios: test prioritization and coverage-guided testing. The experiment evaluation, on three models and four datasets, demonstrates the usefulness and effectiveness of LeCov.

cs.SE↗

arXiv2024

MORTAR: A Model-based Runtime Action Repair Framework for AI-enabled Cyber-Physical Systems

Cyber-Physical Systems (CPSs) are increasingly prevalent across various industrial and daily-life domains, with applications ranging from robotic operations to autonomous driving. With recent advancements in artificial intelligence (AI), learning-based components, especially AI controllers, have become essential in enhancing the functionality and efficiency of CPSs. However, the lack of interpretability in these AI controllers presents challenges to the safety and quality assurance of AI-enabled CPSs (AI-CPSs). Existing methods for improving the safety of AI controllers often involve neural network repair, which requires retraining with additional adversarial examples or access to detailed internal information of the neural network. Hence, these approaches have limited applicability for black-box policies, where only the inputs and outputs are accessible during operation. To overcome this, we propose MORTAR, a runtime action repair framework designed for AI-CPSs in this work. MORTAR begins by constructing a prediction model that forecasts the quality of actions proposed by the AI controller. If an unsafe action is detected, MORTAR then initiates a repair process to correct it. The generation of repaired actions is achieved through an optimization process guided by the safety estimates from the prediction model. We evaluate the effectiveness of MORTAR across various CPS tasks and AI controllers. The results demonstrate that MORTAR can efficiently improve task completion rates of AI controllers under specified safety specifications. Meanwhile, it also maintains minimal computational overhead, ensuring real-time operation of the AI-CPSs.

cs.SE↗

arXiv2024

Controlling the superradiant phase transition in the quantum Rabi model

In the ultrastrong-coupling regime, the quantum Rabi model can exhibit quantum phase transition (QPT) when the ratio of the qubit transition frequency to the frequency of the cavity field approaches infinity. However, it is challenging to control the QPT in few-body systems because of the limited coupling strength and the A^2 term. Here, we propose a practical scheme to manipulate the QPT of the quantum Rabi model in the strong-coupling regime. By applying a periodic frequency modulation to the two-level system in a quantum Rabi model in the strong-coupling regime, an anisotropic quantum Rabi model with ultrastrong and tunable coupling strengths for rotating and counterrotating terms is obtained. The ground state and excitation energy of this model, in terms of the modulation parameters, are studied. We find that the QPT of the quantum Rabi model can be observed in the strong-coupling regime and externally controlled by the modulation.

quant-ph↗