Statistical Frameworks for Uncertainty in Agentic Systems

Modern agentic pipelines are hierarchical compositions of models, tools, and subagents. This workshop studies how rigorous guarantees can support reliable routing, monitoring, stopping, and uncertainty reporting in such systems.

Venue CONF ROOM E1 - E4 COEX Convention & Exhibition Center, Seoul, South Korea
Workshop day July 11, 2026 In person, CONF ROOM E1 - E4
Submissions OpenReview submission site ICML template, 8-page limit, non-archival, deadline April 21, 2026 May 1, 2026*

New!

ICML Conference → Workshop Poster Track

Authors of ICML 2026 accepted papers relevant to uncertainty, calibration, evaluation, risk, reliability, decision-making, or agentic systems are invited to apply to present their paper as a poster at the workshop. We would be glad to have you present your ICML paper as a workshop poster, subject to space constraints.

The workshop is in person on July 11, 2026 in CONF ROOM E1 - E4. Selected papers must be represented by an in-person presenter. There will not be a virtual poster session.

ICML Conference → Workshop Poster Track application deadline: June 10, 2026

ICML Conference → Workshop Poster Track notifications: June 15, 2026

Overview

About the workshop

The workshop aims to connect the fast-moving agentic systems community with the literature on finite-sample and distribution-free guarantees.

It will bring together researchers working on conformal prediction, calibration, prediction-powered inference, sequential testing, and reliable LLM evaluation around a common view of uncertainty in agentic systems.

Themes

  • Distribution-free validity layers for coverage, risk, and abstention under heterogeneity and distribution shift.
  • Anytime-valid sequential inference for continuous monitoring, evidence aggregation, and principled stopping.
  • Uncertainty reporting for interactive components and inter-agent interaction, including auditing and verification when feedback or external services are involved.

Call for Papers

Call for papers

Topics

We invite submissions on the foundations of uncertainty in agentic systems. We welcome theoretical, methodological, and empirical contributions, as well as benchmarks and application-driven case studies.

Topics of interest include, but are not limited to:

  • Conformal prediction and distribution-free risk control for LLM or agent outputs
  • Calibration and uncertainty reporting for generative models
  • Selective prediction, abstention, and hallucination mitigation
  • Sequential monitoring, evidence aggregation, and stopping rules
  • Prediction-powered inference with noisy or machine-generated labels
  • Uncertainty under prompt variation, adaptivity, and distribution shift
  • Auditing and verification for tool use and external feedback loops
  • Compositional guarantees across multi-stage or multi-agent workflows
  • Benchmarks and evaluation protocols for reliable agentic behavior
  • Applications in science, medicine, safety-critical systems, and decision support

Submission details

Submissions should present recent or ongoing work relevant to the workshop themes. The workshop is intended to provide a focused venue for discussion at the intersection of modern agentic systems and rigorous uncertainty quantification.

Submission portal
OpenReview submission site
Submission template
Submissions should use the official ICML template
Camera-ready template
Accepted workshop papers should use the ICML 2026 camera-ready template, with accepted mode enabled by default and the accepted notice adjusted for the AgenticUQ workshop.
Download the ICML 2026 workshop camera-ready template
Camera-ready deadline
June 21, 2026, 23:59 AoE
Page limit
Submissions may be up to 8 pages of main content. Shorter submissions are also welcome.
Archival policy
Submissions are non-archival
Presentation
Accepted workshop papers will appear as posters; selected papers will be invited for contributed spotlight talks
Scope
Previously published, concurrently submitted, and ongoing work may be considered

* Please create your OpenReview profile at least two weeks before the submission deadline.

Important dates

  • Workshop contribution deadline April 21, 2026 May 1, 2026*
  • Notification of acceptance May 16, 2026 AOE
  • ICML Conference → Workshop Poster Track application deadline June 10, 2026
  • ICML Conference → Workshop Poster Track notifications June 15, 2026
  • Accepted workshop paper camera-ready deadline June 21, 2026, 23:59 AoE
  • Workshop day July 11, 2026

Confirmed Speakers

Speakers

Yarin Gal

Yarin Gal

Professor of Machine Learning

University of Oxford

Website
Adam Fisch

Adam Fisch

Research Scientist

Google DeepMind

Website
Ying Jin

Ying Jin

Assistant Professor of Statistics and Data Science

University of Pennsylvania

Profile
Yaniv Romano

Yaniv Romano

Associate Professor of Electrical Engineering and Computer Science

Technion

Website
Andreas Vlachos

Andreas Vlachos

Professor of Natural Language Processing and Machine Learning

University of Cambridge

Website
Seong Joon Oh

Seong Joon Oh

Associate Professor

KAIST AI

Website

Schedule

Schedule

Preliminary structure

The workshop will be held in person on July 11, 2026 in CONF ROOM E1 - E4. The exact timetable will be posted later. The workshop is expected to follow the structure below, with invited talks, contributed spotlights, a panel discussion, and a poster session.

Per the ICML 2026 workshop guidance, workshops typically begin around 8:00 AM and conclude by 5:00 PM.

  • 08:00-08:10Opening remarks
  • MorningInvited talks and contributed talks
  • MiddayBreak
  • AfternoonInvited talks and panel discussion
  • Late afternoonPoster session

Detailed times, titles, and panel composition will be specified later.

Organizers

Organizers

Aymeric Dieuleveut

Aymeric Dieuleveut

Professor of Statistics and Machine Learning

École polytechnique

Website
Maxim Panov

Maxim Panov

Assistant Professor of Machine Learning

MBZUAI

Profile
Stephen Bates

Stephen Bates

Assistant Professor

MIT

Website
Mahmoud Hegazy

Mahmoud Hegazy

PhD Student

École polytechnique and Inria Paris

Website
Aaditya Ramdas

Aaditya Ramdas

Associate Professor

Carnegie Mellon University

Website
Tijana Zrnic

Tijana Zrnic

Assistant Professor

Stanford University

Website