arXiv subjects

Guang Yang

Publications and source records attributed to Guang Yang.

At least 163 records · Page 9Linked to original sources

arXiv2025

Forewarned is Forearmed: Pre-Synthesizing Jailbreak-like Instructions to Enhance LLM Safety Guardrail to Potential Attacks

Despite advances in improving large language model (LLM) to refuse to answer malicious instructions, widely used LLMs remain vulnerable to jailbreak attacks where attackers generate instructions with distributions differing from safety alignment corpora. New attacks expose LLMs' inability to recognize unseen malicious instructions, highlighting a critical distributional mismatch between training data and real-world attacks that forces developers into reactive patching cycles. To tackle this challenge, we propose IMAGINE, a synthesis framework that leverages embedding space distribution analysis to generate jailbreak-like instructions. This approach effectively fills the distributional gap between authentic jailbreak patterns and safety alignment corpora. IMAGINE follows an iterative optimization process that dynamically evolves text generation distributions across iterations, thereby augmenting the coverage of safety alignment data distributions through synthesized data examples. Based on the safety-aligned corpus enhanced through IMAGINE, our framework demonstrates significant decreases in attack success rate on Qwen2.5, Llama3.1, and Llama3.2 without compromising their utility.

cs.CL↗

arXiv2025

MEGA: Spectrophotometric SED Fitting of Little Red Dots Detected in JWST MIRI

We analyze eight spectroscopically confirmed Little Red Dots (LRDs) at redshifts $z = 5.1-8.7$ with JWST/NIRCam, NIRSpec, and MIRI data. The LRDs have red NIRCam colors, F150W-F444W $>$ 1, but flat NIRCam-MIRI colors, $-0.5 < \mathrm{F444W - F770W} < 0.5$, suggesting weak warm/hot dust components. The LRDs have $-1.0 < {F1000W - F1500W} < 1.1$, suggestive of non-uniform rest near-IR properties within the sample. We model the spectral energy distributions (SEDs) of the LRDs using the CIGALE and Prospector codes to assess how the differing templates impact the interpretation for LRDs for cases of: (1) models with star-forming stellar populations only; (2) active galactic nuclei (AGN) dominated models; and (3) composite AGN and star-forming models. Using the Bayesian information criterion, we find that six of the eight LRDs favor AGN models compared to star-forming models, though no model reproduces all of the observed properties. Two LRDs with pronounced Balmer-breaks and broad H$α$ have SEDs that are reproduced with hot, dense-gas ($\log T/\mathrm{K}=5-5.7$, $\log n/\mathrm{cm^{-3}} = 9-11$) models with low dust attenuation ($A(V)\simeq 0.5$ mag). However, these models require an additional thermal component (800-1400 K) to account for the MIRI data, and fail to reproduce the rest-UV and narrow [OIII] emission. The total bolometric emission from the dense-gas models, and possibly CIGALE AGN models, appear consistent with literature constraints in the far-IR and radio, and require $\log L_{bol}/L_\odot<12$. These results suggest that our LRDs cannot be modeled entirely with standard templates, but instead require a novel treatment of gas conditions, AGN and star-formation.

astro-ph.GA↗

arXiv2025

A water Cherenkov detector prototype for future high-energy tau-neutrino experiment

The detection of high-energy tau neutrinos remains a critical challenge in neutrino astronomy, limited by inadequate angular resolution and sensitivity in current detectors like IceCube and KM3NeT. We present a modular water Cherenkov detector prototype optimized for tau-neutrino-induced extensive air showers (EAS) in the 1-100 PeV range, leveraging canyon terrain for natural cosmic-ray shielding. Laboratory validation demonstrates this prototype design has high detection efficiency (>99%) and timing resolution (<2 ns) on MIP particles, enabling precise tau-neutrino-induced EAS reconstruction for future study. The results establish a foundation of a low-cost, scalable neutrino observatory, advancing flavor ratio measurements and cosmic-ray origin problems.

astro-ph.HE↗

arXiv2025

A Physics-Driven Neural Network with Parameter Embedding for Generating Quantitative MR Maps from Weighted Images

We propose a deep learning-based approach that integrates MRI sequence parameters to improve the accuracy and generalizability of quantitative image synthesis from clinical weighted MRI. Our physics-driven neural network embeds MRI sequence parameters -- repetition time (TR), echo time (TE), and inversion time (TI) -- directly into the model via parameter embedding, enabling the network to learn the underlying physical principles of MRI signal formation. The model takes conventional T1-weighted, T2-weighted, and T2-FLAIR images as input and synthesizes T1, T2, and proton density (PD) quantitative maps. Trained on healthy brain MR images, it was evaluated on both internal and external test datasets. The proposed method achieved high performance with PSNR values exceeding 34 dB and SSIM values above 0.92 for all synthesized parameter maps. It outperformed conventional deep learning models in accuracy and robustness, including data with previously unseen brain structures and lesions. Notably, our model accurately synthesized quantitative maps for these unseen pathological regions, highlighting its superior generalization capability. Incorporating MRI sequence parameters via parameter embedding allows the neural network to better learn the physical characteristics of MR signals, significantly enhancing the performance and reliability of quantitative MRI synthesis. This method shows great potential for accelerating qMRI and improving its clinical utility.

eess.IV↗

arXiv2025

Resource-Efficient Automatic Software Vulnerability Assessment via Knowledge Distillation and Particle Swarm Optimization

The increasing complexity of software systems has led to a surge in cybersecurity vulnerabilities, necessitating efficient and scalable solutions for vulnerability assessment. However, the deployment of large pre-trained models in real-world scenarios is hindered by their substantial computational and storage demands. To address this challenge, we propose a novel resource-efficient framework that integrates knowledge distillation and particle swarm optimization to enable automated vulnerability assessment. Our framework employs a two-stage approach: First, particle swarm optimization is utilized to optimize the architecture of a compact student model, balancing computational efficiency and model capacity. Second, knowledge distillation is applied to transfer critical vulnerability assessment knowledge from a large teacher model to the optimized student model. This process significantly reduces the model size while maintaining high performance. Experimental results on an enhanced MegaVul dataset, comprising 12,071 CVSS (Common Vulnerability Scoring System) v3 annotated vulnerabilities, demonstrate the effectiveness of our approach. Our approach achieves a 99.4% reduction in model size while retaining 89.3% of the original model's accuracy. Furthermore, it outperforms state-of-the-art baselines by 1.7% in accuracy with 60% fewer parameters. The framework also reduces training time by 72.1% and architecture search time by 34.88% compared to traditional genetic algorithms.

cs.LG↗

arXiv2025

Conformal Sets in Multiple-Choice Question Answering under Black-Box Settings with Provable Coverage Guarantees

Large Language Models (LLMs) have shown remarkable progress in multiple-choice question answering (MCQA), but their inherent unreliability, such as hallucination and overconfidence, limits their application in high-risk domains. To address this, we propose a frequency-based uncertainty quantification method under black-box settings, leveraging conformal prediction (CP) to ensure provable coverage guarantees. Our approach involves multiple independent samplings of the model's output distribution for each input, with the most frequent sample serving as a reference to calculate predictive entropy (PE). Experimental evaluations across six LLMs and four datasets (MedMCQA, MedQA, MMLU, MMLU-Pro) demonstrate that frequency-based PE outperforms logit-based PE in distinguishing between correct and incorrect predictions, as measured by AUROC. Furthermore, the method effectively controls the empirical miscoverage rate under user-specified risk levels, validating that sampling frequency can serve as a viable substitute for logit-based probabilities in black-box scenarios. This work provides a distribution-free model-agnostic framework for reliable uncertainty quantification in MCQA with guaranteed coverage, enhancing the trustworthiness of LLMs in practical applications.

cs.CL↗

arXiv2025

GEMA-Score: Granular Explainable Multi-Agent Scoring Framework for Radiology Report Evaluation

Automatic medical report generation has the potential to support clinical diagnosis, reduce the workload of radiologists, and demonstrate potential for enhancing diagnostic consistency. However, current evaluation metrics often fail to reflect the clinical reliability of generated reports. Early overlap-based methods focus on textual matches between predicted and ground-truth entities but miss fine-grained clinical details (e.g., anatomical location, severity). Some diagnostic metrics are limited by fixed vocabularies or templates, reducing their ability to capture diverse clinical expressions. LLM-based approaches further lack interpretable reasoning steps, making it hard to assess or trust their behavior in safety-critical settings. These limitations hinder the comprehensive assessment of the reliability of generated reports and pose risks in their selection for clinical use. Therefore, we propose a Granular Explainable Multi-Agent Score (GEMA-Score) in this paper, which conducts both objective quantification and subjective evaluation through a large language model-based multi-agent workflow. Our GEMA-Score parses structured reports and employs stable calculations through interactive exchanges of information among agents to assess disease diagnosis, location, severity, and uncertainty. Additionally, an LLM-based scoring agent evaluates completeness, readability, and clinical terminology while providing explanatory feedback. Extensive experiments validate that GEMA-Score achieves the highest correlation with human expert evaluations on a public dataset, demonstrating its effectiveness in clinical scoring (Kendall coefficient = $0.69$ for ReXVal dataset and Kendall coefficient = $0.45$ for RadEvalX dataset). The anonymous project demo is available at: https://github.com/Zhenxuan-Zhang/GEMA_score.

cs.CL↗

arXiv2025

Hierarchical Structure Sharing Empowers Multi-task Heterogeneous GNNs for Customer Expansion

Customer expansion, i.e., growing a business existing customer base by acquiring new customers, is critical for scaling operations and sustaining the long-term profitability of logistics companies. Although state-of-the-art works model this task as a single-node classification problem under a heterogeneous graph learning framework and achieve good performance, they struggle with extremely positive label sparsity issues in our scenario. Multi-task learning (MTL) offers a promising solution by introducing a correlated, label-rich task to enhance the label-sparse task prediction through knowledge sharing. However, existing MTL methods result in performance degradation because they fail to discriminate task-shared and task-specific structural patterns across tasks. This issue arises from their limited consideration of the inherently complex structure learning process of heterogeneous graph neural networks, which involves the multi-layer aggregation of multi-type relations. To address the challenge, we propose a Structure-Aware Hierarchical Information Sharing Framework (SrucHIS), which explicitly regulates structural information sharing across tasks in logistics customer expansion. SrucHIS breaks down the structure learning phase into multiple stages and introduces sharing mechanisms at each stage, effectively mitigating the influence of task-specific structural patterns during each stage. We evaluate StrucHIS on both private and public datasets, achieving a 51.41% average precision improvement on the private dataset and a 10.52% macro F1 gain on the public dataset. StrucHIS is further deployed at one of the largest logistics companies in China and demonstrates a 41.67% improvement in the success contract-signing rate over existing strategies, generating over 453K new orders within just two months.

cs.LG↗

arXiv2025

Inversion-DPO: Precise and Efficient Post-Training for Diffusion Models

Recent advancements in diffusion models (DMs) have been propelled by alignment methods that post-train models to better conform to human preferences. However, these approaches typically require computation-intensive training of a base model and a reward model, which not only incurs substantial computational overhead but may also compromise model accuracy and training efficiency. To address these limitations, we propose Inversion-DPO, a novel alignment framework that circumvents reward modeling by reformulating Direct Preference Optimization (DPO) with DDIM inversion for DMs. Our method conducts intractable posterior sampling in Diffusion-DPO with the deterministic inversion from winning and losing samples to noise and thus derive a new post-training paradigm. This paradigm eliminates the need for auxiliary reward models or inaccurate appromixation, significantly enhancing both precision and efficiency of training. We apply Inversion-DPO to a basic task of text-to-image generation and a challenging task of compositional image generation. Extensive experiments show substantial performance improvements achieved by Inversion-DPO compared to existing post-training methods and highlight the ability of the trained generative models to generate high-fidelity compositionally coherent images. For the post-training of compostitional image geneation, we curate a paired dataset consisting of 11,140 images with complex structural annotations and comprehensive scores, designed to enhance the compositional capabilities of generative models. Inversion-DPO explores a new avenue for efficient, high-precision alignment in diffusion models, advancing their applicability to complex realistic generation tasks. Our code is available at https://github.com/MIGHTYEZ/Inversion-DPO

cs.CV↗

arXiv2025

ReCoDe: Reinforcement Learning-based Dynamic Constraint Design for Multi-Agent Coordination

Constraint-based optimization is a cornerstone of robotics, enabling the design of controllers that reliably encode task and safety requirements such as collision avoidance or formation adherence. However, handcrafted constraints can fail in multi-agent settings that demand complex coordination. We introduce ReCoDe--Reinforcement-based Constraint Design--a decentralized, hybrid framework that merges the reliability of optimization-based controllers with the adaptability of multi-agent reinforcement learning. Rather than discarding expert controllers, ReCoDe improves them by learning additional, dynamic constraints that capture subtler behaviors, for example, by constraining agent movements to prevent congestion in cluttered scenarios. Through local communication, agents collectively constrain their allowed actions to coordinate more effectively under changing conditions. In this work, we focus on applications of ReCoDe to multi-agent navigation tasks requiring intricate, context-based movements and consensus, where we show that it outperforms purely handcrafted controllers, other hybrid approaches, and standard MARL baselines. We give empirical (real robot) and theoretical evidence that retaining a user-defined controller, even when it is imperfect, is more efficient than learning from scratch, especially because ReCoDe can dynamically change the degree to which it relies on this controller.

cs.RO↗

arXiv2025

Blockchain-Based Decentralized Domain Name System

The current Domain Name System (DNS) infrastructure faces critical vulnerabilities including poisoning attacks, censorship mechanisms, and centralized points of failure that compromise internet freedom and security. Recent incidents such as DNS poisoning attacks on ISP customers highlight the urgent need for resilient alternatives. This paper presents a novel blockchain-based Decentralized Domain Name System (DDNS). We designed a specialized Proof-of-Work blockchain to maximize support for DNS-related protocols and achieve node decentralization. The system integrates our blockchain with IPFS for distributed storage, implements cryptographic primitives for end-to-end trust signatures, and achieves Never Trust, Always Verify zero-trust verification. Our implementation achieves 15-second domain record propagation times, supports 20 standard DNS record types, and provides perpetual free .ddns domains. The system has been deployed across distributed infrastructure in San Jose, Los Angeles, and Orange County, demonstrating practical scalability and resistance to traditional DNS manipulation techniques. Performance evaluation shows the system can handle up to Max Theor. TPS 1,111.1 tx/s (minimal transactions) and Max Theor. TPS 266.7 tx/s (regular transactions) for domain operations while maintaining sub-second query resolution through intelligent caching mechanisms.

cs.CR↗

arXiv2025

MegatronApp: Efficient and Comprehensive Management on Distributed LLM Training

The rapid escalation in the parameter count of large language models (LLMs) has transformed model training from a single-node endeavor into a highly intricate, cross-node activity. While frameworks such as Megatron-LM successfully integrate tensor (TP), pipeline (PP), and data (DP) parallelism to enable trillion-parameter training, they simultaneously expose practitioners to unprecedented systems-level challenges in performance optimization, diagnosis, and interpretability. MegatronApp is an open-source toolchain expressly designed to meet these challenges. It introduces four orthogonal, yet seamlessly composable modules--MegaScan, MegaFBD, MegaDPP, and MegaScope--that collectively elevate the reliability, efficiency, and transparency of production-scale training. This paper presents the motivation, architecture, and distinctive contributions of each module, and elucidates how their synergistic integration augments the Megatron-LM ecosystem.

cs.DC↗

arXiv2025

MindSpeed RL: Distributed Dataflow for Scalable and Efficient RL Training on Ascend NPU Cluster

Reinforcement learning (RL) is a paradigm increasingly used to align large language models. Popular RL algorithms utilize multiple workers and can be modeled as a graph, where each node is the status of a worker and each edge represents dataflow between nodes. Owing to the heavy cross-node dependencies, the RL training system usually suffers from poor cluster scalability and low memory utilization. In this article, we introduce MindSpeed RL, an effective and efficient system for large-scale RL training. Unlike existing centralized methods, MindSpeed RL organizes the essential data dependencies in RL training, i.e., sample flow and resharding flow, from a distributed view. On the one hand, a distributed transfer dock strategy, which sets controllers and warehouses on the basis of the conventional replay buffer, is designed to release the dispatch overhead in the sample flow. A practical allgather--swap strategy is presented to eliminate redundant memory usage in resharding flow. In addition, MindSpeed RL further integrates numerous parallelization strategies and acceleration techniques for systematic optimization. Compared with existing state-of-the-art systems, comprehensive experiments on the RL training of popular Qwen2.5-Dense-7B/32B, Qwen3-MoE-30B, and DeepSeek-R1-MoE-671B show that MindSpeed RL increases the throughput by 1.42 ~ 3.97 times. Finally, we open--source MindSpeed RL and perform all the experiments on a super pod of Ascend with 384 neural processing units (NPUs) to demonstrate the powerful performance and reliability of Ascend.

cs.LG↗

arXiv2025

Impedance-tuned microwave loop for fast, homogeneous Rabi oscillations of a dense ensemble of NV-diamond electronic spins

Obtaining a high Rabi oscillation frequency homogeneously across a spatially-extended population of nitrogen-vacancy (NV) center electronic spins in diamond is useful for efficient spin-state manipulation of the NV ensemble and in using NVs to detect ensembles of other spin species. Here, we achieve a high, homogeneous Rabi frequency for a dense NV ensemble by enhancing the microwave magnetic fields in the center region of a diamond-coupled planar metallic loop via systematic engineering that increases the microwave current driving of the loop, while avoiding off-center proximity to the loop that gives strong but inhomogeneous microwave fields. With such enhanced microwave fields at 2.55 GHz, we achieve a 136.3 MHz NV Rabi frequency with 1.5% inhomogeneity over a 40 $\times$ 40 $μm^{2}$ diamond area; and use the NV ensemble to detect a ~30-MHz magnetic signal, similar to a nuclear magnetic resonance signal at a tesla-scale bias magnetic field, with Hz-scale spectral resolution.

quant-ph↗

arXiv2025

Coupling AI and Citizen Science in Creation of Enhanced Training Dataset for Medical Image Segmentation

Recent advancements in medical imaging and artificial intelligence (AI) have greatly enhanced diagnostic capabilities, but the development of effective deep learning (DL) models is still constrained by the lack of high-quality annotated datasets. The traditional manual annotation process by medical experts is time- and resource-intensive, limiting the scalability of these datasets. In this work, we introduce a robust and versatile framework that combines AI and crowdsourcing to improve both the quality and quantity of medical image datasets across different modalities. Our approach utilises a user-friendly online platform that enables a diverse group of crowd annotators to label medical images efficiently. By integrating the MedSAM segmentation AI with this platform, we accelerate the annotation process while maintaining expert-level quality through an algorithm that merges crowd-labelled images. Additionally, we employ pix2pixGAN, a generative AI model, to expand the training dataset with synthetic images that capture realistic morphological features. These methods are combined into a cohesive framework designed to produce an enhanced dataset, which can serve as a universal pre-processing pipeline to boost the training of any medical deep learning segmentation model. Our results demonstrate that this framework significantly improves model performance, especially when training data is limited.

eess.IV↗

arXiv2025

Lightweight Hypercomplex MRI Reconstruction: A Generalized Kronecker-Parameterized Approach

Magnetic Resonance Imaging (MRI) is crucial for clinical diagnostics but is hindered by prolonged scan times. Current deep learning models enhance MRI reconstruction but are often memory-intensive and unsuitable for resource-limited systems. This paper introduces a lightweight MRI reconstruction model leveraging Kronecker-Parameterized Hypercomplex Neural Networks to achieve high performance with reduced parameters. By integrating Kronecker-based modules, including Kronecker MLP, Kronecker Window Attention, and Kronecker Convolution, the proposed model efficiently extracts spatial features while preserving representational power. We introduce Kronecker U-Net and Kronecker SwinMR, which maintain high reconstruction quality with approximately 50% fewer parameters compared to existing models. Experimental evaluation on the FastMRI dataset demonstrates competitive PSNR, SSIM, and LPIPS metrics, even at high acceleration factors (8x and 16x), with no significant performance drop. Additionally, Kronecker variants exhibit superior generalization and reduced overfitting on limited datasets, facilitating efficient MRI reconstruction on hardware-constrained systems. This approach sets a new benchmark for parameter-efficient medical imaging models.

eess.IV↗

arXiv2025

FreqCross: A Multi-Modal Frequency-Spatial Fusion Network for Robust Detection of Stable Diffusion 3.5 Generated Images

The rapid advancement of diffusion models, particularly Stable Diffusion 3.5, has enabled the generation of highly photorealistic synthetic images that pose significant challenges to existing detection methods. This paper presents FreqCross, a novel multi-modal fusion network that combines spatial RGB features, frequency domain artifacts, and radial energy distribution patterns to achieve robust detection of AI-generated images. Our approach leverages a three-branch architecture: (1) a ResNet-18 backbone for spatial feature extraction, (2) a lightweight CNN for processing 2D FFT magnitude spectra, and (3) a multi-layer perceptron for analyzing radial energy profiles. We introduce a novel radial energy distribution analysis that captures characteristic frequency artifacts inherent in diffusion-generated images, and fuse it with spatial and spectral cues via simple feature concatenation followed by a compact classification head. Extensive experiments on a dataset of 10,000 paired real (MS-COCO) and synthetic (Stable Diffusion 3.5) images demonstrate that FreqCross achieves 97.8\% accuracy, outperforming state-of-the-art baselines by 5.2\%. The frequency analysis further reveals that synthetic images exhibit distinct spectral signatures in the 0.1--0.4 normalised frequency range, providing theoretical foundation for our approach. Code and pre-trained models are publicly available to facilitate reproducible research.

cs.CV↗

arXiv2025

Co-Optimizing Reconfigurable Environments and Policies for Decentralized Multi-Agent Navigation

This work views the multi-agent system and its surrounding environment as a co-evolving system, where the behavior of one affects the other. The goal is to take both agent actions and environment configurations as decision variables, and optimize these two components in a coordinated manner to improve some measure of interest. Towards this end, we consider the problem of decentralized multi-agent navigation in a cluttered environment, where we assume that the layout of the environment is reconfigurable. By introducing two sub-objectives -- multi-agent navigation and environment optimization -- we propose an agent-environment co-optimization problem and develop a coordinated algorithm that alternates between these sub-objectives to search for an optimal synthesis of agent actions and environment configurations; ultimately, improving the navigation performance. Due to the challenge of explicitly modeling the relation between the agents, the environment and their performance therein, we leverage policy gradient to formulate a model-free learning mechanism within the coordinated framework. A formal convergence analysis shows that our coordinated algorithm tracks the local minimum solution of an associated time-varying non-convex optimization problem. Experiments corroborate theoretical findings and show the benefits of co-optimization. Interestingly, the results also indicate that optimized environments can offer structural guidance to de-conflict agents in motion.

cs.RO↗