影响力指数
论文质量、代表作、近期表现、广度与样本量置信度综合计算
63.01/100
前 4%
全站排名 #2,577
发表论文12 篇
平均评分
年均产出4.0 篇/年
Ioannis Mitliagkas
研究方向
Deep learning · Optimization · Machine learning
16
A Derandomization Framework for Structure Discovery: Applications in Neural Networks and Beyond
ICLR 2026Poster
三作16
Beyond Multi-Token Prediction: Pretraining LLMs with Future Summaries
ICLR 2026Poster
37
DisTaC: Conditioning Task Vectors via Distillation for Robust Model Merging
ICLR 2026Poster
10
Unlearning with asymmetric sources: improved unlearning-utility trade-off with public data
ICLR 2026Rejected
通讯18
Solving hidden monotone variational inequalities with surrogate losses
ICLR 2025Poster
9
Compositional Risk Minimization
ICML 2025Poster
35
Understanding Adam Requires Better Rotation Dependent Assumptions
NeurIPS 2025Poster
22
Compositional Risk Minimization
ICLR 2025Rejected
三作14
Performance Control in Early Exiting to Deploy Large Models at the Same Cost of Smaller Ones
ICLR 2025Withdrawn
三作