paper.dinq.meAI 与前沿科学顶会 · 学术新星雷达

更新于

返回新星榜

YM

Youssef Mroueh

ICLR 2026×4AAAI 2026对齐与偏好理论与学习动力学强化学习与后训练推理效率与系统

新星分新星

83

一作贡献60

核心/参与17

方向广度13

合作网络9

2 一作 · 5 篇 · ICLR 2026 / AAAI 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

5

一作ICLR 2026Poster

Information Theoretic Guarantees For Policy Alignment In Large Language Models

Youssef Mroueh · Apoorva Nitsure

对齐与偏好理论与学习动力学OpenReview

作者位次 #5/5ICLR 2026Poster· Theory / Reinforcement Learning and Planning

Best-of-N through the Smoothing Lens: KL Divergence and Regret Analysis

Gholamali Aminian · Idan Shenfeld · Amir Reza Asadi · Ahmad Beirami · Youssef Mroueh

推理强化学习与后训练理论与学习动力学OpenReview

一作ICLR 2026Poster

Revisiting Group Relative Policy Optimization: Insights into On-Policy and Off-Policy Training

Youssef Mroueh · Nicolas Dupuis · Brian Belgodere · Apoorva Nitsure · Mattia Rigotti · Kristjan Greenewald · Jiri Navratil · Jarret Ross · Jesus Rios

强化学习与后训练OpenReview

作者位次 #2/3ICLR 2026Poster· Deep Learning / Algorithms

Guided Speculative Inference for Efficient Test-Time Alignment of LLMs

Jonathan Geuter · Youssef Mroueh · David Alvarez-Melis

效率与系统对齐与偏好OpenReview

作者位次 #4/7AAAI 2026Poster· Artificial Intelligence

GP-MoLFormer-Sim: Test Time Molecular Optimization Through Contextual Similarity Guidance

Jiri Navratil · Jarret Ross · Payel Das · Youssef Mroueh · Samuel C. Hoffman · Vijil Chenthamarakshan · Brian Belgodere

常见合作者

BBBrian Belgodere×2 JNJiri Navratil×2 JRJarret Ross×2 ANApoorva Nitsure GAGholamali Aminian ISIdan Shenfeld AAAmir Reza Asadi ABAhmad Beirami NDNicolas Dupuis ANApoorva Nitsure MRMattia Rigotti KGKristjan Greenewald