1 citations · 1 across the 18 of their papers we have counts for
21 papers
Towards Privacy-Preserving Federated Prompt Tuning under Data Heterogeneity: A Subspace-Decomposed Expert Approach
Yuhua Wang, Xiaodong Li, Yihao Guo +6
Federated prompt tuning (FPT) enables collaborative adaptation of vision--language models (VLMs) using lightweight prompts. Existing methods often address heterogeneity and privacy…
AURORA: Asymmetry and Update-Induced Rotation for Robust Hallucination Detection in Large Language Models
Zishuai Zhang, Hainan Zhang, Zhiming Zheng
Large Language Models (LLMs) have demonstrated remarkable capabilities across a wide range of natural language processing tasks. However, their tendency to generate hallucinations,…
Robust Reasoning via Dynamic Token Selection for Distribution-Aligned Self-Distillation
Ruiqi Zhang, Lingxiang Wang, Hainan Zhang Zhiming Zheng
Self-distillation improves learning efficiency by rewriting reference answers as training data that better matches the model's own distribution. However, reference answers also int…
Replacing Parameters with Preferences: Federated Alignment of Heterogeneous Vision-Language Models
Shule Lu, Yujing Wang, Hainan Zhang +5
Vision-Language Models (VLMs) have broad potential in privacy-sensitive domains such as healthcare and finance, yet strict data-sharing constraints render centralized training infe…
Replacing Parameters with Preferences: Federated Alignment of Heterogeneous Vision-Language Models
Shule Lu, Yujing Wang, Hainan Zhang +5
VLMs have broad potential in privacy-sensitive domains such as healthcare and finance, yet strict data-sharing constraints render centralized training infeasible. FL mitigates this…
From Unfamiliar to Familiar: Detecting Pre-training Data via Gradient Deviations in Large Language Models
Ruiqi Zhang, Lingxiang Wang, Hainan Zhang +2
Pre-training data detection for LLMs is essential for addressing copyright concerns and mitigating benchmark contamination. Existing methods mainly focus on the likelihood-based st…
LLMs Can Unlearn Refusal with Only 1,000 Benign Samples
Yangyang Guo, Ziwei Xu, Si Liu +2
This study reveals a previously unexplored vulnerability in the safety alignment of Large Language Models (LLMs). Existing aligned LLMs predominantly respond to unsafe queries with…
Stable-RAG: Mitigating Retrieval-Permutation-Induced Hallucinations in Retrieval-Augmented Generation
Qianchi Zhang, Hainan Zhang, Liang Pang +2
Retrieval-Augmented Generation (RAG) has become a key paradigm for reducing factual hallucinations in Large Language Models (LLMs), yet little is known about how the order of retri…
Parameter Importance-Driven Continual Learning for Foundation Models
Lingxiang Wang, Hainan Zhang, Zhiming Zheng
Domain-specific post-training often causes catastrophic forgetting, making foundation models lose their general reasoning ability and limiting their adaptability to dynamic real-wo…
An Efficient Framework for Whole-Page Reranking via Single-Modal Supervision
Zishuai Zhang, Sihao Yu, Wenyi Xie +5
The whole-page reranking plays a critical role in shaping the user experience of search engines, which integrates retrieval results from multiple modalities, such as documents, ima…
Can Synthetic Query Rewrites Capture User Intent Better than Humans in Retrieval-Augmented Generation?
JiaYing Zheng, HaiNan Zhang, Liang Pang +2
Multi-turn RAG systems often face queries with colloquial omissions and ambiguous references, posing significant challenges for effective retrieval and generation. Traditional quer…
HyFedRAG: A Federated Retrieval-Augmented Generation Framework for Heterogeneous and Privacy-Sensitive Data
Cheng Qian, Hainan Zhang, Yongxin Tong +2
Centralized RAG pipelines struggle with heterogeneous and privacy-sensitive data, especially in distributed healthcare settings where patient data spans SQL, knowledge graphs, and…
Beyond the Surface: A Solution-Aware Retrieval Model for Competition-level Code Generation
Shiwen Zhang, Lingxiang Wang, Hainan Zhang +3
In competitive programming task, problem statements are often embedded within elaborate narrative backgrounds, requiring deep understanding of the underlying solutions to successfu…
Privacy-Preserving Reasoning with Knowledge-Distilled Parametric Retrieval Augmented Generation
Jinwen Chen, Hainan Zhang, Liang Pang +5
The current RAG system requires uploading plaintext documents to the cloud, risking private data leakage. Parametric RAG (PRAG) encodes documents as LoRA parameters within LLMs, of…
CodeBC: A More Secure Large Language Model for Smart Contract Code Generation in Blockchain
Lingxiang Wang, Hainan Zhang, Qinnan Zhang +4
Large language models (LLMs) excel at generating code from natural language instructions, yet they often lack an understanding of security vulnerabilities. This limitation makes it…
Detecting Stealthy Backdoor Samples based on Intra-class Distance for Large Language Models
Jinwen Chen, Hainan Zhang, Fei Sun +4
Stealthy data poisoning during fine-tuning can backdoor large language models (LLMs), threatening downstream safety. Existing detectors either use classifier-style probability sign…
FedSEA-LLaMA: A Secure, Efficient and Adaptive Federated Splitting Framework for Large Language Models
Zishuai Zhang, Hainan zhang, Weihua Li +4
Private data holds promise for improving LLMs due to its high quality, but its scattered distribution across data silos and the high computational demands of LLMs limit their deplo…
Learning to Erase Private Knowledge from Multi-Documents for Retrieval-Augmented Large Language Models
Yujing Wang, Jinwen Chen, Hainan Zhang +5
Retrieval-Augmented Generation (RAG) is a promising technique for applying LLMs to proprietary domains. However, retrieved documents may contain sensitive knowledge, posing risks o…
HSF: Defending against Jailbreak Attacks with Hidden State Filtering
Cheng Qian, Hainan Zhang, Lei Sha +1
With the growing deployment of LLMs in daily applications like chatbots and content generation, efforts to ensure outputs align with human values and avoid harmful content have int…
AdaComp: Extractive Context Compression with Adaptive Predictor for Retrieval-Augmented Large Language Models
Qianchi Zhang, Hainan Zhang, Liang Pang +2
Retrieved documents containing noise will hinder RAG from detecting answer clues and make the inference process slow and expensive. Therefore, context compression is necessary to e…
MaFeRw: Query Rewriting with Multi-Aspect Feedbacks for Retrieval-Augmented Large Language Models
Yujing Wang, Hainan Zhang, Liang Pang +3
In a real-world RAG system, the current query often involves spoken ellipses and ambiguous references from dialogue contexts, necessitating query rewriting to better describe user'…