World Modeling Workshop

Date: 4-6 February 2026
Location: Agora, Mila - Quebec AI Institute, 6666 Rue Saint-Urbain, Montréal, Canada
Online: Public webinar (freely available to all) (link TBA)

A fundamental desideratum of AI agents is the ability to model environment dynamics and transitions in response to both their own actions and external control signals. This capability, commonly referred to as world modeling (WM), is essential for prediction, planning, and generalization. Learning world models using deep learning has been an active area of research for nearly a decade. In recent years, the field has witnessed significant breakthroughs driven by advances in deep neural architectures and scalable learning paradigms. Multiple subfields, including self-supervised learning (SSL), generative modeling, reinforcement learning (RL), robotics, and large language models (LLMs), have tackled aspects of world modeling, often with different tools and methodologies. While these communities address overlapping challenges, they frequently operate in isolation. As a result, insights and progress in one area may go unnoticed in another, limiting opportunities for synthesis and collaboration. This workshop aims to bridge this gap between subfields of world modeling by fostering open dialogue, critical discussion, and cross-disciplinary exchange. By bringing together researchers from diverse backgrounds, from early-career researchers to established experts, we hope to establish a shared vocabulary, identify common challenges, and surface synergies that can move the field of world modeling forward.

Apply to Attend World Modeling Workshop

We are excited to welcome you to our workshop, whether you are a student, post-doc, professor, or industry professional!

We especially encourage you to share your research by submitting an extended abstract (see Call for Papers). Whether already published or not, consider submitting your work!

Applications with abstract submissions receive priority for registration. Once accepted, you'll receive all the registration details via email. Submit your application here and be part of this unique opportunity.

UPDATE: Registration is closed.

Don't miss the in-person experience! We are hoping to see many of you in-person to learn about your research and to brainstorm as a community!

Student/Post-doc registration fee: 180 CAD

Non-Student/Industry registration fee: 280 CAD

The registration fee includes breakfast, lunch, and coffee breaks!

Call for Papers World Modeling Workshop

Participants interested in presenting a poster are invited to submit an extended abstract with maximum length of 4 pages, excluding references and appendix. Authors can use this opportunity to present early-stage results, insightful negative findings, opinion pieces, or novel datasets. We also welcome published works for submission. All submissions will undergo a double-blinded peer-review process. The accepted abstracts are non-archival.

Please note that at least one of the authors of each submitting paper must register as a reviewer (with the paper submission form). If you are interested in joining the program committee as a reviewer, you can also sign up here!

Please make sure to use the ICLR LaTeX template. For camera ready submissions please refer to the section below (TBA).

Topics of Interest

A non-exhaustive list of relevant topics are:

Self-supervised representations for world modeling
Model-based RL and planning
Generative world modeling
Leveraging world knowledge in LLMs
Embodied AI and Robotics
Causality
World models for scientific discovery
Physical plausibility of world models
World models and AI safety
Datasets and benchmarks for evaluating world models

⚠️ Important Dates 📅

Abstract submission deadline: October 25, 2025 AoE
Paper submission deadline: November 1, 2025 AoE (Submit on OpenReview)
Final decisions to authors: December 1, 2025

Speakers World Modeling Workshop

Keynote

These talks offer broad perspectives on the field of world modeling. They reflect on long-term directions, emerging opportunities, and the potential of world models for advancing AI research over the coming decade.

Polymathic, Flatiron, NYU

Jürgen Schmidhuber

The Swiss AI Lab, KAUST

Amir Zadeh

Lambda

Methods Deep Dive

These talks are in-depth examinations of state-of-the-art methods for world modeling. They focus on the subtle algorithmic choices and practical implementation details that critically influence the quality of world models.

Florian Bordes (Meta)

IntPhys 2

Karen Ullrich & Mark Ibrahim (Meta)

OpenApps

Bayan Bruss (Capital One)

Time and Money

Stephen Spencer (GDM)

Genie

Tutorials

These tutorials are hands-on sessions that provide background on core topics and ingredients of world modeling, offer practical design choices, and guide participants through implementation.

Randall Balestriero
(Brown University, Meta)

Self-Supervised Learning

Glen Berseth
(Mila, UdeM)

Vision-Language-Action Models

Alessandro Lazaric (Meta)

Reinforcement Learning

Schedule (times in EST) World Modeling Workshop

Day 1

Wednesday, 04-02-2026

8:00	Registration and Breakfast
8:50	Opening Remarks
9:00	Keynote 1 (Remote): Jürgen Schmidhuber Title: TBA
10:00	Keynote 2: Sherry Yang Title: From pretraining world models to policy improvement
11:00	Coffee Break
11:15	Keynote 3: Shirley Ho Title: Polymathic AI: A path towards scientific discovery
12:15	Lunch Break
13:20	Group Photo
13:30	Keynote 4: Yoshua Bengio Title: Bayesian modeling of causal structure for AI safety
14:30	Keynote 5: Yann LeCun Title: TBA
15:30	Coffee Break
15:40	Panel Discussion Panelists: Yoshua Bengio, Yann LeCun, Sherry Yang, Shirley Ho, Alessandro Lazaric Moderator: Randall Balestriero
16:40	Methods Deep Dive 1: Mido Assran
17:25	Poster Session 1 and Networking

Day 2

Thursday, 05-02-2026

8:30	Breakfast
9:00	Keynote 6: Amir Zadeh Title: World Modeling: Infrastructure and Vision at Lambda
10:00	Methods Deep Dive 2: Stephen Spencer
10:45	Coffee Break
11:00	Oral Presentations 1 (3 papers)
11:45	Methods Deep Dive 3: Bayan Bruss
12:30	Lunch Break
13:45	Methods Deep Dive 4: Florian Bordes
14:30	Oral Presentations 2 (4 papers)
15:30	Coffee Break
15:45	Methods Deep Dive 5: Zhiting Hu
16:30	Methods Deep Dive 6: Mark Ibrahim and Karen Ullrich
17:15	Poster Session 2 and Networking

Day 3

Friday, 06-02-2026

8:30	Breakfast
9:00	Tutorial 1: RL (Alessandro Lazaric)
11:15	Coffee Break
11:30	Tutorial 2: SSL (Randall Balestriero)
12:30	Lunch Break
13:30	Tutorial 2 (continued)
14:45	Coffee Break
15:00	Tutorial 3: VLA (Glen Berseth)
17:15	Closing Remarks

Practical Information World Modeling Workshop

Stay In Touch

We encourage all participants to join our Slack channel to receive workshop announcements and connect with other attendees. We will share timely updates there, including schedule changes, room directions, and any additional practical logistics.

Slack Invitation Link: Click here to join the workspace

Transportation

> Transport from YUL (Montréal–Trudeau) Airport to Montréal Downtown

[Option 1] Uber / Taxi / Rideshare (Recommended)

Taxis and rideshare (e.g., Uber) are available directly at the airport. Follow airport signage to the official taxi stand or the rideshare pickup area. This is typically the simplest option if you have luggage or arrive late.

[Option 2] 747 Airport Bus (public transit)

The STM 747 bus connects YUL airport to downtown and metro connections. It is a reliable option during most hours of the day. Tickets are purchased at the airport (valid on buses/metro in Zone A). More info on the STM website.

> Transport from YUL (Montréal–Trudeau) Airport to Mila (workshop venue)

Fastest and simplest

Take a taxi or rideshare from YUL directly to Mila (Québec AI Institute) 6650 Rue Saint-Urbain, Montréal, QC H2S 3H1.

Public transit route (airport → metro → Mila)

1. Take the 747 bus from YUL.

2. Get off at the stop "Berri-UQAM" (about 45 minutes from the airport).

3. Connect to the Montréal metro (station Berry-UQAM).

4. Take the Orange line direction “Montmorency” and get off at Beaubien station.

5. Walk from Beaubien station to Mila (about 10–15 minutes, depending on your pace).

> Transport from Montréal Downtown to Mila

By metro (recommended)

1. Take the Orange line direction “Montmorency”.

2. Exit at Beaubien station.

3. Walk to Mila (Québec AI Institute) 6650 Rue Saint-Urbain, Montréal, QC H2S 3H1. (about 10–15 minutes).

Metro ticket / fare

You need a valid STM fare to enter the metro (Zone A). You can buy tickets at any metro station from the vending machines (credit/debit accepted). Common options include: Single trip (one ride), 2 trips or day pass (1-day / 24-hour). More info on the STM website

By uber/taxi

Taxis and rideshare are also convenient from downtown, especially outside peak commuting times. Take in account potential traffic delays due to rush hours or weather conditions.

Workshop Location

The workshop will take place in Mila - Agora (6650 Rue Saint-Urbain, Montréal, QC H2S 3H1). After entering the building, follow the internal signage to “Agora”. Event staff will also be present near the entrance during peak arrival times to help direct participants. Mila Entrance

Hotel Recommendation

Please note: unfortunately, we do not have any preferential rate, discount code, or reserved room block with hotels. We recommend booking as soon as possible, as availability may vary.

Recommended hotels (downtown area):

- Fairmont The Queen Elizabeth
- Montreal Marriott Château Champlain
- Le Centre Sheraton Montréal
- Hotel Bonaventure Montréal
- Hyatt Place Montréal – Downtown

Tip: If possible, choose a hotel near a metro station to simplify commuting.

Montréal Weather in February

Montréal in February is known to be quite cold 🥶, with average temperatures ranging from -10°C (14°F) to -2°C (28°F). It is advisable to dress warmly in layers, including a good winter coat, gloves, a hat, and insulated footwear. Snow is common during this time of year, so be prepared for winter weather conditions.

Organizers World Modeling Workshop