CoRL 2026 Workshop

Modeling Uncertainty in Robotic World Models

Date
November 12, 2026
Location
Austin, Texas, USA
Venue
10th Conference on Robot Learning
Submission deadline
September 30, 2026 AoE
Submit a paper Full call for papers Contact organizers

Abstract

Recent advances in vision-language foundation models and video generation have made learned world models an increasingly important tool for robotic task planning, policy simulation, and reward modeling. Yet real-world robots must operate under uncertainty that many current world models do not adequately represent: noisy sensors, partial observability, imprecise actuation, distribution shift, and unpredictable humans or other agents.

Although related communities have studied uncertainty through belief-space planning, probabilistic robotics, conformal prediction, goal inference, safe autonomy, and uncertainty quantification, it remains unclear how these ideas should scale to the high-dimensional latent spaces used by modern learned world models. This workshop will bring together researchers from robot learning, planning, probabilistic robotics, safe autonomy, uncertainty quantification, and human-robot interaction to address that gap.

We aim to clarify which uncertainty representations fit which robotic settings, how uncertainty propagates into downstream planning and reinforcement learning, and which benchmarks and metrics are needed to measure progress. Through invited talks, poster presentations, and roundtable discussions, our goal is to move robotic world models beyond expressiveness alone toward reliable uncertainty awareness for deployment in unstructured real-world environments.

Important Dates

{{ d.label }}
{{ d.value }}

Call for Papers

{{ g.label }}
{{ p.t }}

Areas of Interest

Topics include, but are not limited to:

{{ t.n }} {{ t.text }}

Invited Speakers

{{ s.n }}
{{ s.name }}
{{ s.affiliation }}

Roundtable speakers will be announced.

Confirmed Schedule

Room and location: TBD. All times local to Austin (CST).

{{ s.time }}
{{ s.event }}
Massachusetts Institute of Technology Harvard University Princeton University University of Pennsylvania Brown University New York University NVIDIA

Advisory Committee

{{ p.name }}
{{ p.affiliation }}

Organizing Committee

{{ p.name }}
{{ p.affiliation }}

Accepted Papers

39 accepted papers.

Detecting Silent Action-Interface Mismatch with Calibrated Relational World Models

Rui Zhu, xurui li, Haixu Tang

CalIm: Learning a Calibrated Uncertainty Expectation for World-Model Imagination

Zhejia Wen, Meixia Lin, Kangjie Zhou, Pengying Wu, Chunxuan Jiao, Chang Liu, Lige Liu

Beyond Ensemble Disagreement: Predictive STL Monitoring of RL Locomotion under Distribution Shift

Qurban Ali, Badaruddin Chachar, Maqsood Ali Solangi

Relational Invariance under Predictive Disagreement in World Models

Weichun Tai, Yih Ping Luh

Adaptive Rollout Truncation Based on Epistemic Uncertainty for Efficient Offline World Model Training

Nikodem Sebastian Zymla, Laurin Thiele, Johannes Pitz

Latent Projection for Latent Planning

Yoo Sung Jang, Michael S Ryoo

Decision-Level Auditing of Latent Robot World Models

Anikait Rana, Siddhant Rana, Aditya Rao Udupi

ManiDreams: An Open-Source Library for Robust Object Manipulation via Uncertainty-aware Task-specific Intuitive Physics

Gaotian Wang, Kaiyu Hang, Kejia Ren, Andrew Morgan

Most of a World Model's Epistemic Uncertainty Is Not Worth Resolving

Varun Puttagunta

When Can Learned Uncertainty Be Trusted? An Audit of Distributional Reinforcement Learning

Hari Prasad

MS-MEM: Multi-Skill Manipulation-Enhanced Mapping via Uncertainty-Aware Belief-Space World Models

Nils Dengler, Yitian Shi, Jesper Mücke, Sicong Pan, Rania Rayyes, Maren Bennewitz

Closure Atlas: Effective Closure and Its Failure Boundaries in World Models

Xingrui Gu, Chuyi Jiang, Hanxue Gu, Felix Juefei-Xu, Mengyue Yang

When Uncertainty Should Change the Plan: Adequacy Floors and Defeaters for Robotic World Models

Ken Howarth

Hallucination in World Models is Predictable and Preventable

Nicklas Hansen, Xiaolong Wang

Unified Belief-Conditioned World Modeling and Parameter Estimation for Multi-Finger Manipulation

Abhinav Kumar, Soshi Iba, Rana Soltani Zarrin, Dmitry Berenson

Training-Free Uncertainty Maps for Detecting Unreliable Rollouts in Driving World Models

Joseph Bruno, Denis Liu, Javier Yu, Charles C Ochoa

AUC-CoT: Action-Uncertainty-Conditioned Chain-of-Thought Reasoning for Vision-Language-Action Models

Sehee Kweon, Wooyul Jung, Geunhee Kim, Donghwa Kang, Jiwon Mun, Ali Pesaranghader, Alexander W Olson, Frank C. Park, Byoung-Tak Zhang

Latent World Models Enhance Imitation Learning with Foresight

Hengkai Pan, Gaoyue Zhou, Yann LeCun, Lerrel Pinto, Deepak Pathak

Test Time Augmentation and Learned Variance for Heteroscedastic Uncertainty in LiDAR Scene flow

Daniel Jean-Soo Rhee, Leilani H. Gilpin

Toward Uncertainty-Aware, Physics-Grounded Human World Models

Yufei Zhang, Zijun Cui

Noise or Ignorance? Auditing the Uncertainty Signals of Ensemble World Models against Known Noise and Error

Tianlang Chen

Do Household World Models Know When Their Memories Are Stale?

Oliver Wang

Energy-Based Policies Can Detect Their Own Failures

Jalal Naghiyev, Kirill Khvastow, Alexi Gladstone, Ralf Römer, Haonan Chen, Heng Ji, Yilun Du, Angela P. Schoellig

Localized Conformal Safety Monitoring with Vision-Language Models for Autonomous Driving

Luís Marques, Rong Fang, Disha Kamale, Dmitry Berenson

PhysiCalWorld: In-Context Physics Calibration for World Models

Yunhao Luo, Wentao He, Yinpei Dai, Hyelin Nam, Hongze Fu, Joyce Chai, Nima Fazeli, Jeong Joon Park

Modeling Latent Disturbances for Robust Decision-Making in World Models

Junwon Seo, Andrea Bajcsy

StressDream: Steering Video World Models for Robust Policy Evaluation and Improvement

Junwon Seo, Sushant Veer, Ran Tian, Wenhao Ding, Apoorva Sharma, Karen Leung, Edward Schmerling, Marco Pavone, Andrea Bajcsy

Articulated Object Manipulation with Uncertainty-Aware Kinematic Modeling

Ruiyao Liu, Boshu Lei, Zhuoyang Pan, Kostas Daniilidis

Wiggle and Go! System Identification for Zero-Shot Dynamic Rope Manipulation

Arthur Jakobsson, Abhinav Mahajan, Karthik Pullalarevu, Krishna Suresh, Yunchao Yao, Yuemin Mao, Bardienus Pieter Duisterhof, Shahram Najam Syed, Jeffrey Ichnowski

Beyond AND and OR: Shaping the No-Alarm Region for Failure Detection in Robotics

Laura Lützow, Ahmad A. Rushdi, Mykel Kochenderfer

Partially Observable Object Rearrangement via Task-Centric Active Perception

Sahit Chintalapudi, Xiaoyan Wang, Aditya Agarwal, Tomás Lozano-Pérez, Leslie Pack Kaelbling

Knowing Where to Look: Interactive Object Search via Spatial-Semantic Fusion

Aditya Agarwal, Lucy Cai, Kelly Lu, Travis Manderson, Nicholas Roy, Tomás Lozano-Pérez, Leslie Pack Kaelbling

World Models Under Asynchronous Sensor Observations

Akash Anand, Abhay Anand, Yash Vishe

Does Best-of-K World-Model Selection Help a Diffusion Policy Succeed?

Ahmad A. Rushdi, Mykel Kochenderfer

MBCE: TD-Guided Action Editing for Offline Model-Based RL

Jisang Ahn

Reads Right, Judges Wrong: Near-Miss Failures and the Limits of Uncertainty Recovery in VLM Success Detectors

Hamza Naeem

Knowing When Not to Answer: Measurability Gates Forecasts, Not Decisions

Brownstafford Abraham

FOCUS: Flow-Matching with Offline Critical-Interaction Update Supervision for Robot World Models

Weiyi Hong, Chenjun Bai

JACK: Joint Axis Correction from Kinesthetic Feedback

Arthur Jakobsson, Eugene Lee, Bardienus Pieter Duisterhof, Huang Huang, Jeff Tan, Romina Mir, Jean Oh, Jeffrey Ichnowski