作者位次 #3/4ICLR 2026Poster· Theory / Probabilistic Methods
Softmax is not Enough (for Adaptive Conformal Classification)
Navid Akhavan Attar · Hesam Asadollahzadeh · Ling Luo · Uwe Aickelin
理论与学习动力学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。
Navid Akhavan Attar · Hesam Asadollahzadeh · Ling Luo · Uwe Aickelin
Jie Yang · Ling Luo · Nestor Cabello · Lars Kulik
Recent research has shown that single-vector embedding retrieval models face a fundamental bottleneck: the finite dimensionality of single-vector representations limits their capacity to represent arbitrary top-k relevant item combinations, even with perfect training. This inherent bottleneck sub...
Xuan-May Le · Minh-Tuan Tran · Ling Luo · Uwe Aickelin · Dinh Q. Phung · Trung Le