paper.dinq.meAI 与前沿科学顶会 · 学术新星雷达

更新于

返回新星榜

SK

Sham Kakade

ICLR 2026×6强化学习与后训练检索与知识效率与系统可解释性

新星分贡献者

56

一作贡献0

核心/参与25

方向广度8

合作网络12

0 一作 · 6 篇 · ICLR 2026

获取完整研究档案

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。

联系方式

实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。

2026 顶会论文

6

作者位次 #3/4ICLR 2026Poster· Optimization / Everything Else

The Potential of Second-Order Optimization for LLMs: A Study with Full Gauss-Newton

Natalie Abreu · Nikhil Vyas · Sham Kakade · Depen Morwani

作者位次 #4/5ICLR 2026Poster· Deep Learning / Generative Models and Autoencoders

In Good GRACES: Principled Teacher Selection for Knowledge Distillation

Abhishek Panigrahi · Bingbin Liu · Sadhika Malladi · Sham Kakade · Surbhi Goel

强化学习与后训练检索与知识OpenReview

作者位次 #5/8ICLR 2026Poster

Any-Order Flexible Length Masked Diffusion

Jaeyeon Kim · Lee Kit · Carles Domingo i Enrich · Yilun Du · Sham Kakade · Timothy Ngotiaoco · Sitan Chen · Michael Albergo

作者位次 #3/5ICLR 2026Poster· Reinforcement Learning / Deep RL

Parameter-Efficient Reinforcement Learning using Prefix Optimization

Itamar Rocha Filho · Rosie Zhao · Sham Kakade · Eran Malach · Samy Jelassi

强化学习与后训练效率与系统OpenReview

作者位次 #4/7ICLR 2026Poster

Cognitive models can reveal interpretable value trade-offs in language models

Sonia Murthy · Rosie Zhao · Jennifer Hu · Sham Kakade · Markus Wulfmeier · Peng Qian · Tomer Ullman

可解释性OpenReview

作者位次 #6/6ICLR 2026Poster

Seesaw: Accelerating Training by Balancing Batch Size and Learning Rate Scheduling

Alexandru Meterez · Depen Morwani · Jingfeng Wu · Costin-Andrei Oncescu · Cengiz Pehlevan · Sham Kakade

常见合作者

DMDepen Morwani×2 RZRosie Zhao×2 NANatalie Abreu NVNikhil Vyas APAbhishek Panigrahi BLBingbin Liu SMSadhika Malladi SGSurbhi Goel JKJaeyeon Kim LKLee Kit CECarles Domingo i Enrich YDYilun Du