I used to teach AI & ML for media, art & design: PerttuHamalainen/MediaAI: Aalto University's Intelligent Computational Media course (AI & ML for media, art & design)
However, I killed the course as one can increasingly get better results with commercial tools than tweaking and finetuning models oneself. Instead, I’m now focusing more on teaching advanced game programming and technical art, with some AI & ML material sprinkled in.
Meanwhile, I’ll keep posting significant new papers etc. here.
Note that this section is very much under construction, waiting for me to clean up and migrate stuff from my private iCloud notes.
Table of contents
February 25, 2026
New research following Anthropic’s seminal work in the domain:
https://arxiv.org/abs/2512.01797
Abstract:
Large language models (LLMs) frequently generate hallucinations – plausible but factually incorrect outputs – undermining their reliability. While prior work has examined hallucinations from macroscopic perspectives such as training data and objectives, the underlying neuron-level mechanisms remain largely unexplored.
In this paper, we conduct a systematic investigation into hallucination-associated neurons (H-Neurons) in LLMs from three perspectives: identification, behavioral impact, and origins. Regarding their identification, we demonstrate that a remark-ably sparse subset of neurons (less than 0.1% of total neurons) can reliably predict hallucination occurrences, with strong generalization across diverse scenarios.
In terms of behavioral impact, controlled interventions reveal that these neurons are causally linked to over-compliance behaviors. Concerning their origins, we trace these neurons back to the pre-trained base models and find that these neurons remain predictive for hallucination detection, indicating they emerge during pre-training.