The Fifth International Workshop on Spatio-Temporal Reasoning and Learning (STRL 2026) will take place in Bremen, Germany, collocated with IJCAI 2026 (https://2026.ijcai.org/), the premier international gathering of researchers in AI!

Website: https://strl-workshop.github.io/strl2026/



About the Workshop
============

Spatio-temporal reasoning is becoming a central bottleneck for modern AI. Across language, vision, robotics, and multi-agent settings, current learning-based systems can be impressive at pattern recognition, yet they often fail when faced with distribution shifts, long-horizon dependencies, counterfactual queries, or the need to justify decisions in dynamic environments. At the same time, the community is seeing rapid progress in foundation models (including vision-language-action models), neural world models, and spatio-temporal representation learning, which creates an opportunity to couple these advances with structured representations and reasoning to improve robustness, transfer, and explainability.

STRL 2026 brings together the Knowledge Representation and Reasoning and Machine Learning communities to advance dependable spatio-temporal AI. We invite submissions that develop representations, learning methods, and evaluation resources for reasoning about space, time, motion, and events in dynamic environments (see topics below). A key goal is to identify reusable abstractions and benchmarks that improve robustness, generalization, and explainability across tasks and modalities.



Call for Papers
=========

Topics of Interest
-------------------

We invite submissions related to the integration of spatial and temporal reasoning with machine learning, including but not limited to:

-Neuro-symbolic approaches to spatio-temporal reasoning and learning
-Spatio-temporal commonsense reasoning and learning
-Generalization in spatio-temporal reasoning and learning (compositional generalization, continual learning, transfer learning, causal reasoning)
-Large language models and vision-language(-action) models for spatio-temporal reasoning and learning
-World models for spatio-temporal reasoning and learning
-Explainability and interpretability in spatio-temporal reasoning and learning
-Spatio-temporal reasoning and learning in adjacent fields (e.g., robotics, computer vision, natural language processing, multi-agent systems, reinforcement learning, and knowledge graphs)
-Benchmarks for spatio-temporal reasoning and learning
-Real-world problems in spatio-temporal reasoning and learning

Keywords: Spatio-Temporal Reasoning, Neuro-Symbolic AI, Multimodal Learning, Commonsense Reasoning, World Models, Explainable AI, Benchmarks and Evaluation, Continual and Compositional Generalization

Submission Guidelines
-------------------------

We welcome submissions across the full spectrum of theoretical and practical work including research ideas, methods, applications, tools, benchmarks, surveys, and position papers.

Papers should be formatted according to the IJCAI-ECAI conference guidelines (The updated LaTeX styles and Word template are available at https://www.ijcai.org/authors_kit).

Short papers (4 pages, excluding references and appendices)
Regular papers (8 pages, excluding references and appendices)
Extended abstracts (2 pages, excluding references and appendices)
Unlimited additional pages containing references and supplementary materials are allowed.

The workshop is non-archival, i.e., authors are allowed to submit work that is under review or will be submitted elsewhere. We welcome ongoing and unpublished work. We will also accept papers that are under review at the time of submission, or that have been recently accepted without published proceedings. Extended abstracts of published papers are welcome, as long as they demonstrate a close link to the topic of the workshop. Authors of selected papers may be invited to submit an extended version of their work to a journal special issue related to the workshop.

Use of LLMs in Submissions: While we encourage the use of LLMs for brainstorming and drafting, we require that all submissions reflect the authors’ original work and ideas. Therefore, any submissions that include errors due to LLM-generated content (e.g., hallucinated references, inaccurate information) will be desk-rejected without review.

All papers will be peer-reviewed in a double-blind process and assessed based on their novelty, technical quality, potential impact, clarity, and reproducibility (when applicable). At least one author per submission must commit to reviewing for the workshop. You will need to designate the reviewing author(s) on charingtool. Submissions without a nominated reviewer may be desk-rejected.

Submission website: https://chairingtool.com/conferences/strl2026/main-track?role=author



Organization
========

Organizing Committee
-------------------------

Jae Hee Lee, University of Hamburg, Germany
Jakob Suchan, Constructor University Bremen, Germany
Xun Gong, Southwest Jiaotong University, China

Advisory Committee
-----------------------

Michael Sioutis, University of Montpellier, France
Zhiguo Long, Southwest Jiaotong University, China
Parisa Kordjamshidi, Michigan State University, US
Mehul Bhatt, Örebro University, Sweden