3 citations · 7 across the 29 of their papers we have counts for
10 papers · 1 filter
ARENA: Automated Red-Teaming for Large Audio Language Models
Jiaming He, Zhicong Huang, Tian Jin +5
Large audio-language models (LALMs) make it possible to interact with language models through speech, music, and environmental sound, but they also introduce a safety surface that…
Trigger the Straggler: Load Hijack on Mixture-of-Experts LLMs
Rui Zhang, Wenbo Jiang, Hongwei Li +4
Expert parallelism (EP) is a common strategy for serving large Mixture-of-Experts (MoE) models across multiple GPUs by distributing experts among devices. Router decisions then det…
EvoTrustRAG: Evolution-Aware Conflict Attribution and Evidence Handling for Reliable Retrieval-Augmented Generation
Xi Nie, Hongwei Li, Shenghao Wu +3
Retrieval-Augmented Generation (RAG) improves the factuality of large language models with external knowledge, yet conflicting evidence remains a fundamental challenge in dynamic a…
InkShield: Writing Style Protection Against Unauthorized Handwriting Mimicry
Jian Xiong, Wenbo Jiang, Zihan Wang +4
Recent handwritten text generators can reproduce a writer's style from publicly available references, posing risks of document forgery and identity misuse. An attacker may use a pu…
When Poison Fails After Retrieval: Revisiting Corpus Poisoning under Chunking and Reranking Pipelines
Xi Nie, Hongwei Li, Shenghao Wu +3
Retrieval-Augmented Generation (RAG) systems are vulnerable to corpus poisoning attacks that manipulate downstream model outputs through malicious knowledge injection. Existing stu…
The Art of (Mis)alignment: How Fine-Tuning Methods Effectively Misalign and Realign LLMs in Post-Training
Rui Zhang, Hongwei Li, Yun Shen +6
The deployment of large language models (LLMs) raises significant ethical and safety concerns. While LLM alignment techniques are adopted to improve model safety and trustworthines…
UTOPIA: Unlearnable Tabular Data via Decoupled Shortcut Embedding
Jiaming He, Fuming Luo, Hongwei Li +5
Unlearnable examples (UE) have emerged as a practical mechanism to prevent unauthorized model training on private vision data, while extending this protection to tabular data is no…
BadTemplate: A Training-Free Backdoor Attack via Chat Template Against Large Language Models
Zihan Wang, Hongwei Li, Rui Zhang +2
Chat template is a common technique used in the training and inference stages of Large Language Models (LLMs). It can transform input and output data into role-based and templated…
Semantic-level Backdoor Attack against Text-to-Image Diffusion Models
Tianxin Chen, Wenbo Jiang, Hongqiao Chen +2
Text-to-image (T2I) diffusion models are widely adopted for their strong generative capabilities, yet remain vulnerable to backdoor attacks. Existing attacks typically rely on fixe…
State Backdoor: Towards Stealthy Real-world Poisoning Attack on Vision-Language-Action Model in State Space
Ji Guo, Wenbo Jiang, Yansong Lin +6
Vision-Language-Action (VLA) models are widely deployed in safety-critical embodied AI applications such as robotics. However, their complex multimodal interactions also expose new…