5 papers
Mitigating Database Leakage in RAG Systems with Keyword-Grounded Fact Substitution
Ziliang Zhang, Yubo Zhu, Wei Tong +4
Retrieval-Augmented Generation (RAG) has emerged as a powerful paradigm for combining large language models (LLMs) with external knowledge sources. However, RAG systems remain vuln…
PURPOSE: Poisoning Conflict Resolution in RAG via Proxy-Fact-Grounded Updates
Zijian Wang, Yubo Zhu, Muzhi Dong +7
In Retrieval-Augmented Generation (RAG), post-retrieval conflict resolution arbitrates among noisy or contradictory retrieved passages. However, the robustness of this safeguard ag…
Auditing Data Provenance in LLM Fine-tuning via Intrinsic Distributional Fingerprints
Zirui Huang, Yunlong Mao, Wei Tong +3
The proliferation of customized Large Language Models (LLMs) poses critical risks of Data Intellectual Property (Data IP) infringement via unauthorized fine-tuning on proprietary d…
The LLM Already Knows: Estimating LLM-Perceived Question Difficulty via Hidden Representations
Yubo Zhu, Dongrui Liu, Zecheng Lin +3
Estimating the difficulty of input questions as perceived by large language models (LLMs) is essential for accurate performance evaluation and adaptive inference. Existing methods…
On Evaluating the Poisoning Robustness of Federated Learning under Local Differential Privacy
Zijian Wang, Wei Tong, Tingxuan Han +4
Federated learning (FL) combined with local differential privacy (LDP) enables privacy-preserving model training across decentralized data sources. However, the decentralized data-…