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
41.56/100
前 13.3%
全站排名 #8,587
发表论文7 篇
平均评分
年均产出3.5 篇/年
Alexandre Rame
研究方向
Ensembling and weight averaging · Out-of-distribution and domain generalization · Deep learning · Reinforcement learning from human preferences
18
BOND: Aligning LLMs with Best-of-N Distillation
ICLR 2025Poster
23
Diversity-Rewarded CFG Distillation
ICLR 2025Poster
通讯16
Streaming DiLoCo with overlapping communication
COLM 2025Poster
18
WARP: On the Benefits of Weight Averaged Rewarded Policies
ICLR 2025Rejected
一作10
On Teacher Hacking in Language Model Distillation
ICML 2025Poster