作者位次 #6/6ICLR 2026Poster· Deep Learning / Robustness
WRING Out The Bias: A Rotation-Based Alternative To Projection Debiasing
Walter Gerych · Cassandra Parent · Quinn Perian · Rafiya Javed · Justin Solomon · Marzyeh Ghassemi
社会与人本OpenReview
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。
Walter Gerych · Cassandra Parent · Quinn Perian · Rafiya Javed · Justin Solomon · Marzyeh Ghassemi
Yuxin Xiao · Sana Tonekaboni · Walter Gerych · Vinith Suriyakumar · Marzyeh Ghassemi
Kimia Hamidieh · Veronika Thost · Walter Gerych · Mikhail Yurochkin · Marzyeh Ghassemi
Yuxin Xiao · Shujian Zhang · Marzyeh Ghassemi · Wenxuan Zhou
To acquire instruction-following capabilities, large language models (LLMs) undergo instruction tuning, where they are trained on instruction-response pairs using next-token prediction (NTP). Efforts to improve instruction tuning often focus on higher-quality supervised fine-tuning (SFT) datasets...