作者位次 #5/7ICML 2026Poster· Social Aspects / Fairness
Knowing Bias, Doing Better: Mitigating Social Bias in LLMs via Know-Bias Neuron Enhancement
Jinhao Pan · Chahat Raj · Anjishnu Mukherjee · Sina Mansouri · Bowen Wei · Shloka Yada · Ziwei Zhu
首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。
实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。
Jinhao Pan · Chahat Raj · Anjishnu Mukherjee · Sina Mansouri · Bowen Wei · Shloka Yada · Ziwei Zhu
Chahat Raj · Bowen Wei · Aylin Caliskan · Antonios Anastasopoulos · Ziwei Zhu
While bias in large language models (LLMs) is well-studied, similar concerns in vision-language models (VLMs) have received comparatively less attention. Existing VLM bias studies often focus on portrait-style images and gender-occupation associations, overlooking broader and more complex social...
Mehrdad Fazli · Bowen Wei · Ziwei Zhu
Hallucinations—generating responses inconsistent with the visual input—remain a critical limitation of large vision-language models (LVLMs), especially in open-ended tasks such as image captioning and visual reasoning. In this work, we probe the layer-wise generation dynamics that drive hallucina...
Bowen Wei · Mehrdad Fazli · Ziwei Zhu