Zum Inhalt springen
Technische Universität München

Master-Seminar on World Models

Overview

In this course students will learn about recent advances in the field of world models. They will study relevant articles related to topics on world models and present them in the form of a scientific presentation. The course covers a broad range of topics related to world models such as, generative models and their conditioning, sequence prediction of probabilistic dynamic systems, multi-modal embeddings of observations and knowledge and other related topics. The course will cover articles from a wide range of top-tier conferences in machine learning, computer vision and robotics and various application areas.

Pre-Course Meeting

The pre-course meeting was held on January 30th at 01:30 pm. You can find the slides here.

Registration

The registration for this course runs through the TUM matching system

To increase your chances you are encouraged to submit a motivation letter to:

swm(at)camp.cit.tum.de

providing the follow information

  • Name and email
  • Motivation to take the course
  • Previous experiences (courses, projects, …)
  • Latest CV (not mandatory)
  • Transcripts of records (not mandatory)

The deadline for the submissions is 17.02.2026.

Master-Seminar on World Models (IN2107)

Lecturer (assistant)
Nassir Navab [L]
Diego Biagini
Nikolas Brasch
Changxuan Li
Kunyi Li
Nils Morbitzer
Klara Reichard
Federico Tombari
Ning Wang
Number 0000001762
Type seminar
Duration 2 SWS
Term Sommersemester 2026
Language of instruction English
Position within curricula See TUMonline
Dates See TUMonline

Dates

Admission information

See TUMonline
Note: Please use the matching system for the course registration (http://docmatching.in.tum.de). Interested students should attend the pre-course meeting. Details can be found on the course website (https://www.cs.cit.tum.de/camp/teaching/).

Objectives

The students will learn about recent methods on world models and related topics. They will learn how to independently research a topic with the help of scientific literature and how to prepare and present a scientific presentation.

Description

In this course students will learn about recent advances in the field of world models. They will study relevant articles related to topics on world models and present them in the form of a scientific presentation. The course covers a broad range of topics related to world models such as, generative models and their conditioning, sequence prediction of probabilistic dynamic systems, multi-modal embeddings of observations and knowledge and other related topics. The course will cover articles from a wide range of top-tier conferences in machine learning, computer vision and robotics and various application areas.

Prerequisites

Fundamental knowledge in machine learning and deep learning are required. Advanced knowledge on vision, language and generative models is beneficial.

Teaching and learning methods

The students will work independently and dive deep into an individual topic. Each student is supervised by a tutor familiar with the research topic to ask questions to, discuss methodology with and receive feedback from throughout the course. The student will learn about a wider range of topics through the presentations of the other students.

Examination

The students will be evaluated based on their literature research and the preparation and presentation of their scientific talk.

Recommended literature

The website https://github.com/knightnemo/Awesome-World-Models provides a list of recent works on world models as well as overview articles.

Links

To top