Research Blog
08.05.2026
Right Answer, Wrong Reasoning - Is AI Thinking or Cheating?
MCML Research Insight – With Xinpeng Wang and Barbara Plank
Imagine a student solving a math problem. The steps look perfectly logical, neatly written, and convincing. But in reality, the student already knew the answer and simply worked backwards to justify …
27.04.2026
Research Stay at Imperial College London
Lennart Bastian – Funded by the MCML AI X-Change Program
During my research stay at Imperial College London, I had the pleasure of working with the CIRCLE group led by Tolga Birdal in the Department of Computing. Among the various research topics at CIRCLE …
23.04.2026
When Vision AI Hallucinates Details
MCML Research Insight – With Rui Xiao, Sanghwan Kim and Zeynep Akata
A photo shows a dog on a sofa. You ask an AI assistant: Is there also a cat sitting next to it? The assistant confidently says yes, even though there is no cat at all. In many everyday situations, …
16.04.2026
Do Language Models Reason Like Humans?
MCML Research Insight – With Philipp Mondorf and Barbara Plank
Imagine reading the sentence: “If it rains, the streets will be wet.” Most people would consider it perfectly reasonable. Now consider a different statement: “If the moon is made of cheese, the …
02.04.2026
How AI Avatars Shape Perceived Fairness
MCML Research Insight – With Ka Hei Carrie Lau and Enkelejda Kasneci
Imagine applying for a job and being interviewed not by a human recruiter, but by an AI avatar on your screen. The conversation feels surprisingly natural. The interviewer smiles, asks questions, and …
19.03.2026
Teaching Models to Say ‘I’m Not Sure’
MCML Research Insight – With Hannah Laus, Felix Krahmer and Holger Rauhut
Machine learning models are good at giving answers, predicting patterns, and classifying objects but they are much worse at saying when they are unsure. But what if the model could admit when it was …
05.03.2026
Foundations of Diffusion: One Map for Images and Text
MCML Research Insight – With Vincent Pauline, Tobias Höppe, Andrea Dittadi, and Stefan Bauer
From hyper-realistic video generation to protein design, Diffusion Models are the engine behind the current wave of Generative AI. But if you try to understand how diffusion models actually work, you …
19.02.2026
COSMOS – Teaching Vision-Language Models to Look Beyond the Obvious
MCML Research Insight - With Sanghwan Kim, Rui Xiao, Mariana-Iuliana Georgescu, Stephan Alaniz, and Zeynep Akata
Today’s AI systems are remarkably good at recognizing what stands out most. Yet understanding how details relate to each other remains surprisingly difficult. Imagine asking an AI assistant to …
05.02.2026
Needle in a Haystack: Finding Exact Moments in Long Videos
MCML Research Insight - With With Tanveer Hannan and Thomas Seidl
Long videos are everywhere, with footage of movies, YouTube videos, body-cam recordings, and AR/VR often running for tens of minutes or even hours. Now imagine asking a simple question like “Where are …
2024-11-22 - Last modified: 2026-05-11