Hui Liu
7 indexed papers
Publications per year
Top categories
Frequent co-authors
Research Timeline
The EvoSafety framework enhances LLM safety by externalizing attack and defense mechanisms, enabling persistent, transferable, and model-agnostic robustness against adversarial prompts.
The paper introduces DSL-LLaDA, a method that lightly adapts a pre-trained masked diffusion language model to perform continuous denoising in embedding space, significantly improving text generation quality and robustness, especially under low step budgets.
The paper introduces D3IM, a novel parameter-free sampler that enables direct revision of visible tokens in Masked Diffusion Language Models, and proposes SCOPE to mitigate the model's tendency to perpetuate errors.
The paper introduces DSL-LLaDA, a method that lightly adapts a pre-trained masked diffusion language model to perform continuous denoising in embedding space, significantly improving text generation quality and robustness, especially under low step budgets.
The paper introduces D3IM, a novel parameter-free sampler that enables direct revision of visible tokens in Masked Diffusion Language Models, and proposes SCOPE to mitigate the model's tendency to perpetuate errors.
This paper investigates the vulnerability of LLM-based automatic grading systems to prompt injection (PI) attacks, demonstrating that current systems are highly susceptible to manipulation that can lead to unfairly high scores.
The paper proposes Conditional Retrieval Alignment (CoRA), a gradient-free framework for task-conditioned retrieval using a frozen encoder and paired inputs and outputs.
Papers
Gradient-free Task-Conditioned Retrieval for On-Device In-Context Learning
Xinyu Luo, Hui Liu, Yihua Shao, Junyi Yang +2 more
The paper proposes Conditional Retrieval Alignment (CoRA), a gradient-free framework for task-conditioned retrieval using a frozen encoder and paired inputs and outputs.