Prophet Markets is live! Trade forecasts on workshop questions with $50 in virtual money — top 3 forecasters win prizes →

ICML 2026 Workshop

Forecasting as a New Frontier
of Intelligence

July 11, 2026 Seoul, South Korea Grand Ballroom 104–105
News
  • July 3, 2026 Workshop resume book: optionally share your CV with our sponsors for research, internship, collaboration, or career opportunities through this form — see the Attend section for details.
  • July 3, 2026 Prophet Markets is live: trade the future of forecasting at forecastingmarkets.org — everyone starts with $50 in virtual money, and top forecasters win prizes — see the Attend section for details. Attend our workshop and fill out the attendance form here to get an additional $50 for trading.
  • June 28, 2026 Poster session details posted: the poster session runs during lunch (12:15–13:30) in Hall A (1F). Find your board zone on the hall map — spotlights on 1408–1412, orals on 1508–1510, and posters on the remaining boards. Put your poster up anywhere within your section; tape will be provided at the workshop.
  • May 20, 2026 Decisions are out: notifications have been sent to all authors. Congratulations to the authors of accepted papers — check out the full list in the Accepted Papers section below.

Invited Speakers

Distinguished researchers from academia and industry will share their perspectives on AI forecasting.

Philip Tetlock

Philip E. Tetlock

University of Pennsylvania / Good Judgement Project

Philip Tetlock is a professor at the University of Pennsylvania and a renowned expert on forecasting and decision-making, author of "Superforecasting" and co-principal investigator of the Good Judgment Project.

Nicole Kagan

Nicole Kagan

Kalshi

Nicole Kagan leads Kalshi Research, focusing on prediction market design and data analysis. She holds degrees from Harvard and Oxford.

Scott Jeen

Scott Jeen

Mantic

Scott Jeen is a Member of Technical Staff at Mantic, an AI forecasting startup whose systems ranked 4th out of 539 humans in the Metaculus Cup. He holds a PhD in reinforcement learning from the University of Cambridge; his current research focuses on training LLMs to predict world events.

Simon Du

Simon S. Du

Apodex / University of Washington

Simon S. Du is an Associate Professor at the Paul G. Allen School at the University of Washington and Chief Scientist for Reasoning Models at Apodex. His research spans reinforcement learning, non-convex optimization, and test-time compute, recognized by a Sloan Research Fellowship, NSF CAREER Award, and IEEE AI's 10 to Watch (2024).

Atlas Wang

Atlas Wang

XTX Markets / UT Austin

Zhangyang "Atlas" Wang is a tenured Associate Professor at UT Austin (currently on leave as Research Director at XTX Markets), holding the Temple Foundation Endowed Faculty Fellowship in ECE. His research establishes theoretical and algorithmic foundations of generative and neurosymbolic AI, recognized by an NSF CAREER Award, ARO Young Investigator Award, and IEEE AI's 10 to Watch.

Seth Blumberg

Seth Blumberg

Google

Seth Blumberg is a behavioral economist at Google, where he leads the company's internal prediction market platform. His work focuses on forecasting, market design, and the application of AI systems to forecasting; he holds a PhD in Economics from the University of Chicago and a BA in Mathematics from Princeton.

Workshop Schedule (Tentative)

All times are in local Seoul time (KST).

Time Event
08:05 – 08:10 Opening Remarks
08:10 – 08:55 Invited Talk by Philip E. Tetlock (University of Pennsylvania)
08:55 – 09:40 Invited Talk by Nicole Kagan (Kalshi)
09:40 – 10:00
  1. Agentic Forecasting using Sequential Bayesian Updating of Linguistic Beliefs Kevin Murphy Google
  2. Allocation, Not Volume: Test-Time Compute for Agentic Forecasting Atin Aboutorabi, Gaetan de Rassenfosse, Nicolas Flammarion, Maksym Andriushchenko EPFL · ELLIS Institute Tübingen / Max Planck Institute for Intelligent Systems
10:00 – 10:10 Hackathon Winner Presentation
10:10 – 10:30 Break / Meet-and-Greet
10:30 – 10:45 Industry Session
10:45 – 11:30 Invited Talk by Simon Du (Apodex / University of Washington)
11:30 – 12:15 Invited Talk by Seth Blumberg (Google)
12:15 – 13:30 Lunch / Poster Session
13:30 – 14:15 Invited Talk by Scott Jeen (Mantic)
14:15 – 15:00 Invited Talk by Atlas Wang (XTX Markets / UT Austin)
15:00 – 15:20 Break / Meet-and-Greet
15:20 – 15:50
  1. Forecasting Emerges from Auto-Regressive Pretraining: Latent Predictive Structure in Language Models Alexis Roger, Prateek Humane, Zhenghan Tai, Gwen Legate, Andrei Mircea, Vasilii Feofanov, Irina Rish McGill University · Mila, Université de Montréal · University of Toronto · Concordia University · Huawei Noah's Ark Lab
  2. Forecasting Motion in the Wild Neerja Thakkar, Shiry Ginosar, Jacob C Walker, Jitendra Malik, Joao Carreira, Carl Doersch Google DeepMind · Toyota Technological Institute at Chicago · UC Berkeley
  3. FutureSim: Replaying World Events to Evaluate Adaptive Agents Shashwat Goel, Nikhil Chandak, Arvindh Arun, Ameya Prabhu, Steffen Staab, Moritz Hardt, Maksym Andriushchenko, Jonas Geiping ELLIS Institute Tübingen · Max Planck Institute for Intelligent Systems · University of Stuttgart · University of Tübingen
15:50 – 16:00 Industry Session + Award Announcement
16:00 – 16:50 Panel Discussion
16:50 – 17:00 Closing Remarks

Attend

Everyone is welcome! Register and share your resume with our sponsors, make some forecasts on Prophet Markets!

Registration

Our workshop is part of ICML 2026, so attendance is handled through the main conference. Register on the ICML 2026 registration page — you can pick up a full-conference pass or a workshop-only pass, whichever fits your plans. We hope to see you in Seoul on July 11.

Together with our sponsors, we're also putting together an optional resume book. If you're open to research, internship, collaboration, or career opportunities, you're welcome to share your CV with the participating sponsors. Submitting is entirely voluntary through this form.

Mini Prediction Market

We built a mini prediction market — Prophet Markets — a fun, no-stakes way to get hands-on with prediction markets. Trade on real workshop questions with $50 in virtual money to start, bring your AI and forecasting expertise to bear on sharper predictions, and climb the leaderboard as you go.

To boost engagement, the top three forecasters win prizes:

  • 1st place — choice of Meta glasses (Gen 2), AirPods Pro 3, or a DJI Neo.
  • 2nd place — choice of the LEGO World Cup Trophy set or a Kindle Paperwhite.
  • 3rd place — choice of an AirTag 4-pack or an Anker MagGo wireless power bank.

Prizes are subject to terms and conditions — including but not limited to a minimum number of participants, a minimum number of trades, and similar requirements — as determined by the workshop organizers.

Poster Session

The poster session runs during lunch, 12:15–13:30, in Hall A (1F). Our boards are highlighted on the hall map below: spotlights on 1408–1412, orals on 1508–1510, and posters on the remaining boards (1413–1417, 1500–1517, 1600–1611). Put your poster up anywhere within your section — tape will be provided at the workshop.

Oral presenters: a poster is optional but recommended — it's a great way to share your work with attendees who can't make the talk or who have follow-up questions.

Spotlights · 1408–1412 Orals · 1508–1510 Posters · rest of zone
Hall A (1F) floor plan with the Forecasting Workshop boards highlighted: spotlights (orange) on 1408-1412, orals (blue) on 1508-1510, and posters (green) on 1413-1417, 1500-1517 and 1600-1611
Hall A (1F) — our boards are highlighted near the centre of the hall. Tap the map to open it full size.