Guozhi Liu — accepted AI-conference papers

2026

Experiential Fairness: Bridging the Gap Between User Experience and Resource-Centric Fairness in Online LLM Services

Jiahua Huang, Wentai Wu, Yongheng Liu, Guozhi Liu, Yang Wang, Weiwei Lin

AAAI 2026technical

Conventional fairness in multi-tenant Large Language Model (LLM) inference services is typically defined by system-centric metrics such as equitable resource allocation. We argue that this is unilateral and it creates a gap between measured system performance and actual user-perceived quality. We ch

Cited by 0 Source ↗PDF ↗Scholar ↗

2026

Surgery: Mitigating Harmful Fine-Tuning for Large Language Models via Attention Sink

Guozhi Liu, Weiwei Lin, Tiansheng Huang, Ruichao Mo, Qi Mu, Xiumin Wang +1

ICML 2026spotlight

Harmful fine-tuning can invalidate safety alignment of large language models, exposing significant safety risks. In this paper, we utilize the attention sink mechanism to mitigate harmful fine-tuning. Specifically, we first measure a statistic named *sink divergence* for each attention head and obse…

Cited by 0 Source ↗Scholar ↗