作者位次 #5/9ACL 2026Poster· Natural Language Processing
C ontext L ens: Modeling Imperfect Privacy and Safety Context for Legal Compliance
Haoran Li · Yulin Chen · Huihao Jing · Wenbin Hu · Tsz Ho Li · Chanhou Lou · Hong Ting Tsang · Sirui Han · Yangqiu Song
摘要
Individuals’ concerns about data privacy and AI safety are highly contextualized and extend beyond sensitive patterns. Addressing these issues requires reasoning about the context to identify and mitigate potential risks. Though researchers have widely explored using large language models (LLMs)...