Songda Li — accepted AI-conference papers

2024

Better Late Than Never: Model-Agnostic Hallucination Post-Processing Framework Towards Clinical Text Summarization

Songda Li, Yunqi Zhang, Chunyuan Deng, Yake Niu, Hui Zhao

ACL 2024findings

Clinical text summarization has proven successful in generating concise and coherent summaries. However, these summaries may include unintended text with hallucinations, which can mislead clinicians and patients. Existing methods for mitigating hallucinations can be categorized into task-specific an…

Cited by 1 Source ↗PDF ↗code ↗Scholar ↗

2024

Think Before You Act: A Two-Stage Framework for Mitigating Gender Bias Towards Vision-Language Tasks

Yunqi Zhang, Songda Li, Chunyuan Deng, Luyi Wang, Hui Zhao

NAACL 2024long

Gender bias in vision-language models (VLMs) can reinforce harmful stereotypes and discrimination. In this paper, we focus on mitigating gender bias towards vision-language tasks. We identify object hallucination as the essence of gender bias in VLMs. Existing VLMs tend to focus on salient or famili…

Cited by 1 Source ↗PDF ↗code ↗Scholar ↗