paper.dinq.meAI 与前沿科学顶会 · 学术新星雷达

更新于

返回新星榜

AB

Amrit Singh Bedi

ICML 2026×4IROS 2026×3AAAI 2026强化学习与后训练社会与人本安全智能体与工具多模态

新星分资深

41

一作贡献0

核心/参与39

方向广度16

合作网络9

0 一作 · 8 篇 · ICML 2026 / IROS 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

8

作者位次 #4/4ICML 2026Poster· Social Aspects / Alignment

The Hidden Cost of Structured Generation in LLMs: Draft-Conditioned Constrained Decoding

Avinash Reddy · Thayne Walker · Jaime Ide · Amrit Singh Bedi

对齐与偏好代码社会与人本OpenReview

作者位次 #2/2ICML 2026Poster· Social Aspects / Everything Else

Position: We Need AI Efficiency Incentives for Accessibility and Sustainability

Marco Bornstein · Amrit Singh Bedi

效率与系统社会与人本OpenReview

作者位次 #6/6ICML 2026Poster

Safety Recovery in Reasoning Models Is Only a Few Early Steering Steps Away

Soumya Suvra Ghosal · Souradip Chakraborty · Vaibhav Singh · Furong Huang · Dinesh Manocha · Amrit Singh Bedi

推理安全可解释性OpenReview

作者位次 #4/4ICML 2026Poster· Deep Learning / Large Language Models

RL with Learnable Textual Feedback: A Bilevel Approach

Utsav Singh · Sidhaarth Murali · Souradip Chakraborty · Amrit Singh Bedi

强化学习与后训练OpenReview

作者位次 #4/4IROS 2026Oral· Adaptive Allocation of Computational Resources

Training Fast Robot Policies with Slow Foundation Models

Utsav Singh · Pramit Bhattacharyya · Vinay Namboodiri · Amrit Singh Bedi

强化学习与后训练Reinforcement LearningDeep Learning MethodsMachine Learning for Robot ControlOpenReview

作者位次 #5/6IROS 2026Oral· Finding Vulnerabilities in Autonomous Systems

SABER: A Stealthy Agentic Black-Box Attack Framework for Vision-Language-Action Models

Xiyang Wu · Guangyao Shi · Qingzi Wang · Zongxia Li · Amrit Singh Bedi · Dinesh Manocha

安全智能体与工具多模态Robot SafetyAgent-Based SystemsReinforcement LearningOpenReview

作者位次 #2/8IROS 2026Oral· The Mechanics of Multi-Robot Teamwork

Learning Multi-Robot Coordination through Locality-Based Factorized Multi-Agent Actor-Critic Algorithm

Chak Lam Shek · Amrit Singh Bedi · Anjon Basak · Ellen Novoseller · Nicholas Waytowich · Priya Narayanan · Dinesh Manocha · Pratap Tokekar

智能体与工具强化学习与后训练检索与知识Cooperating RobotsReinforcement LearningMulti-Robot SystemsOpenReview

作者位次 #4/5AAAI 2026Poster· Artificial Intelligence

SafeR-CLIP: Mitigating NSFW Content in Vision-Language Models While Preserving Pre-Trained Knowledge

Adeel Yousaf · Joseph Fioresi · James Beetham · Amrit Singh Bedi · Mubarak Shah

多模态数据与预训练检索与知识OpenReview

常见合作者

SCSouradip Chakraborty×2 USUtsav Singh×2 DMDinesh Manocha×2 ARAvinash Reddy TWThayne Walker JIJaime Ide MBMarco Bornstein SGSoumya Suvra Ghosal VSVaibhav Singh FHFurong Huang DMDinesh Manocha SMSidhaarth Murali