Johannes Mehrer

Johannes Mehrer

Postdoctoral researcher in computational neuroscience at EPFL's NeuroAI Lab. I study brain-like computation using deep neural networks and model-guided causal interventions.

Dreaming out loud: A self-synthesis approach for training vision-language models with developmentally plausible data

Published in BabyLM Challenge at Conference on Computational Natural Language Learning (CoNLL), 2024

Recommended citation: AlKhamissi, Tang, Gokce, Mehrer*, Schrimpf* (*shared last authorship). (2024). "Dreaming out loud: A self-synthesis approach for training vision-language models with developmentally plausible data." CoNLL 2024.
Download Paper