4 papers
When LoRA Betrays: Backdooring Text-to-Image Models by Masquerading as Benign Adapters
Liangwei Lyu, Jiaqi Xu, Jianwei Ding +1
Low-Rank Adaptation (LoRA) has emerged as a leading technique for efficiently fine-tuning text-to-image diffusion models, and its widespread adoption on open-source platforms has f…
A Bayesian Framework for Human-AI Collaboration: Complementarity and Correlation Neglect
Saurabh Amin, Amine Bennouna, Daniel Huttenlocher +3
We develop a decision-theoretic model of human-AI interaction to study when AI assistance improves or impairs human decision-making. A human decision-maker observes private informa…
Wikipedia Contributions in the Wake of ChatGPT
Liang Lyu, James Siderius, Hannah Li +3
How has Wikipedia activity changed for articles with content similar to ChatGPT following its introduction? We estimate the impact using differences-in-differences models, with dis…
MisinfoEval: Generative AI in the Era of "Alternative Facts"
Saadia Gabriel, Liang Lyu, James Siderius +3
The spread of misinformation on social media platforms threatens democratic processes, contributes to massive economic losses, and endangers public health. Many efforts to address…