Yongfeng Huang
YOU?
Author Swipe
View article: Higher-order spectral form factors of circular unitary ensemble
Higher-order spectral form factors of circular unitary ensemble Open
Spectral form factor (SFF), one of the key quantity from random matrix theory, serves as an important tool to probe universality in disordered quantum systems and quantum chaos. In this work, we present exact closed-form expressions for th…
View article: SN-Stego: Dataset for Social Networks Text Steganalysis via Local Group Discovery and Sample Distribution Regulation
SN-Stego: Dataset for Social Networks Text Steganalysis via Local Group Discovery and Sample Distribution Regulation Open
View article: Phase transitions driven by solute concentration, temperature, and pressure in uranium-6wt % niobium alloy
Phase transitions driven by solute concentration, temperature, and pressure in uranium-6wt % niobium alloy Open
An angular-dependent potential for the U-Nb system is developed based on an existing ADP for U and a new EAM potential for Nb through fitting flexible cross-interaction functions and alloy parameters to experimental and first-principles da…
View article: Extraction of lithium from cookeite-bearing claystone by chlorination roasting: From thermodynamic calculation to experimental practice
Extraction of lithium from cookeite-bearing claystone by chlorination roasting: From thermodynamic calculation to experimental practice Open
View article: TaxAgent: How Large Language Model Designs Fiscal Policy
TaxAgent: How Large Language Model Designs Fiscal Policy Open
Economic inequality is a global challenge, intensifying disparities in education, healthcare, and social stability. Traditional systems like the U.S. federal income tax reduce inequality but lack adaptability. Although models like the Saez…
View article: SWE-SQL: Illuminating LLM Pathways to Solve User SQL Issues in Real-World Applications
SWE-SQL: Illuminating LLM Pathways to Solve User SQL Issues in Real-World Applications Open
Resolution of complex SQL issues persists as a significant bottleneck in real-world database applications. Current Large Language Models (LLMs), while adept at text-to-SQL translation, have not been rigorously evaluated on the more challen…
View article: Reinforcement Learning-based Copyright Protection Watermarking for Large Language Model
Reinforcement Learning-based Copyright Protection Watermarking for Large Language Model Open
View article: Clustering-Driven Pseudo-Labeling in Source-Free Domain Adaptation for Linguistic Steganalysis
Clustering-Driven Pseudo-Labeling in Source-Free Domain Adaptation for Linguistic Steganalysis Open
View article: MrM: Black-Box Membership Inference Attacks against Multimodal RAG Systems
MrM: Black-Box Membership Inference Attacks against Multimodal RAG Systems Open
Multimodal retrieval-augmented generation (RAG) systems enhance large vision-language models by integrating cross-modal knowledge, enabling their increasing adoption across real-world multimodal tasks. These knowledge databases may contain…
View article: Enhancing Watermarking Quality for LLMs via Contextual Generation States Awareness
Enhancing Watermarking Quality for LLMs via Contextual Generation States Awareness Open
Recent advancements in watermarking techniques have enabled the embedding of secret messages into AI-generated text (AIGT), serving as an important mechanism for AIGT detection. Existing methods typically interfere with the generation proc…
View article: HeteRAG: A Heterogeneous Retrieval-augmented Generation Framework with Decoupled Knowledge Representations
HeteRAG: A Heterogeneous Retrieval-augmented Generation Framework with Decoupled Knowledge Representations Open
Retrieval-augmented generation (RAG) methods can enhance the performance of LLMs by incorporating retrieved knowledge chunks into the generation process. In general, the retrieval and generation steps usually have different requirements fo…
View article: pFedGPA: Diffusion-based Generative Parameter Aggregation for Personalized Federated Learning
pFedGPA: Diffusion-based Generative Parameter Aggregation for Personalized Federated Learning Open
Federated Learning (FL) offers a decentralized approach to model training, where data remains local and only model parameters are shared between the clients and the central server. Traditional methods, such as Federated Averaging (FedAvg),…
View article: Learning Human Feedback from Large Language Models for Content Quality-aware Recommendation
Learning Human Feedback from Large Language Models for Content Quality-aware Recommendation Open
Recommender systems are widely employed to mitigate information overload by tailoring online content to individual preferences. Existing recommendation methods typically focus on optimizing the relevance between candidate item content and …
View article: Efficient Diffusion Models: A Survey
Efficient Diffusion Models: A Survey Open
Diffusion models have emerged as powerful generative models capable of producing high-quality contents such as images, videos, and audio, demonstrating their potential to revolutionize digital content creation. However, these capabilities …
View article: Preface: Security and safety for next generation industrial systems
Preface: Security and safety for next generation industrial systems Open
View article: Shifting-Merging: Secure, High-Capacity and Efficient Steganography via Large Language Models
Shifting-Merging: Secure, High-Capacity and Efficient Steganography via Large Language Models Open
In the face of escalating surveillance and censorship within the cyberspace, the sanctity of personal privacy has come under siege, necessitating the development of steganography, which offers a way to securely hide messages within innocen…
View article: Long-context Language Models Fail in Basic Retrieval Tasks Without Sufficient Reasoning Steps
Long-context Language Models Fail in Basic Retrieval Tasks Without Sufficient Reasoning Steps Open
View article: Roles of Pre-Existing Dislocation Network on the Shock Responses of Single Crystal Tantalum
Roles of Pre-Existing Dislocation Network on the Shock Responses of Single Crystal Tantalum Open
View article: Mitigate Position Bias in LLMs via Scaling a Single Hidden States Channel
Mitigate Position Bias in LLMs via Scaling a Single Hidden States Channel Open
View article: Semantic Steganography: A Framework for Robust and High-Capacity Information Hiding using Large Language Models
Semantic Steganography: A Framework for Robust and High-Capacity Information Hiding using Large Language Models Open
In the era of Large Language Models (LLMs), generative linguistic steganography has become a prevalent technique for hiding information within model-generated texts. However, traditional steganography methods struggle to effectively align …
View article: “Water-in-montmorillonite” quasi-solid-state electrolyte for ultralow self-discharge aqueous zinc-ion batteries
“Water-in-montmorillonite” quasi-solid-state electrolyte for ultralow self-discharge aqueous zinc-ion batteries Open
The practical application of aqueous zinc-ion batteries (AZIBs) is limited by zinc dendrites, parasitic reactions, and self-discharging. Quasi-solid-state electrolytes (QSSEs) are promising solutions but have high costs, low conductivity, …
View article: pFedGPA: Diffusion-based Generative Parameter Aggregation for Personalized Federated Learning
pFedGPA: Diffusion-based Generative Parameter Aggregation for Personalized Federated Learning Open
Federated Learning (FL) offers a decentralized approach to model training, where data remains local and only model parameters are shared between the clients and the central server. Traditional methods, such as Federated Averaging (FedAvg),…
View article: GLoCIM: Global-view Long Chain Interest Modeling for news recommendation
GLoCIM: Global-view Long Chain Interest Modeling for news recommendation Open
Accurately recommending candidate news articles to users has always been the core challenge of news recommendation system. News recommendations often require modeling of user interest to match candidate news. Recent efforts have primarily …
View article: SETTP: Style Extraction and Tunable Inference via Dual-level Transferable Prompt Learning
SETTP: Style Extraction and Tunable Inference via Dual-level Transferable Prompt Learning Open
Text style transfer, an important research direction in natural language processing, aims to adapt the text to various preferences but often faces challenges with limited resources. In this work, we introduce a novel method termed Style Ex…
View article: Provably Robust and Secure Steganography in Asymmetric Resource Scenario
Provably Robust and Secure Steganography in Asymmetric Resource Scenario Open
To circumvent the unbridled and ever-encroaching surveillance and censorship in cyberspace, steganography has garnered attention for its ability to hide private information in innocent-looking carriers. Current provably secure steganograph…
View article: Co-Stega: Collaborative Linguistic Steganography for the Low Capacity Challenge in Social Media
Co-Stega: Collaborative Linguistic Steganography for the Low Capacity Challenge in Social Media Open
View article: Mitigate Position Bias in Large Language Models via Scaling a Single Dimension
Mitigate Position Bias in Large Language Models via Scaling a Single Dimension Open
Large Language Models (LLMs) are increasingly applied in various real-world scenarios due to their excellent generalization capabilities and robust generative abilities. However, they exhibit position bias, also known as "lost in the middl…
View article: Towards Next-Generation Steganalysis: LLMs Unleash the Power of Detecting Steganography
Towards Next-Generation Steganalysis: LLMs Unleash the Power of Detecting Steganography Open
Linguistic steganography provides convenient implementation to hide messages, particularly with the emergence of AI generation technology. The potential abuse of this technology raises security concerns within societies, calling for powerf…
View article: Learnable Linguistic Watermarks for Tracing Model Extraction Attacks on Large Language Models
Learnable Linguistic Watermarks for Tracing Model Extraction Attacks on Large Language Models Open
In the rapidly evolving domain of artificial intelligence, safeguarding the intellectual property of Large Language Models (LLMs) is increasingly crucial. Current watermarking techniques against model extraction attacks, which rely on sign…
View article: Towards the Robustness of Differentially Private Federated Learning
Towards the Robustness of Differentially Private Federated Learning Open
Robustness and privacy protection are two important factors of trustworthy federated learning (FL). Existing FL works usually secure data privacy by perturbing local model gradients via the differential privacy (DP) technique, or defend ag…