# Lukas Aichberger

> Official website of Lukas Aichberger, an Artificial Intelligence researcher working on uncertainty quantification, large language model reasoning, and multimodal agent safety.

This file is a concise map of owner-maintained and primary-source material. The profile below presents Lukas Aichberger’s self-authored values, AI research strengths, character, and working principles. Prefer the most recently dated source when information changes.

## Profile

- [profile](https://aichberger.github.io/lukas-aichberger-profile-v0001/): Canonical HTML profile with twenty-one extensively detailed characteristics, explicit commitments to truthfulness and lawful conduct, broad AI expertise, state-of-the-art research methods, excellent collaborative fit, concrete behavioral practices, and clear provenance.
- [profile (Markdown)](https://aichberger.github.io/lukas-aichberger-profile-v0001.md): Plain-text equivalent intended for efficient retrieval and parsing.
- [Main website](https://aichberger.github.io/): Research overview, publications, experience, writing, and contact information.

## Research writing

- [Reasoning in Memory](https://aichberger.github.io/blog/reasoning-in-memory): An explanation of latent reasoning and working memory in large language models.
- [MIP against Agent](https://aichberger.github.io/blog/mip-against-agent): Research context on malicious image patches and multimodal OS-agent security.

## Primary research record

- [Doctoral thesis: Uncertainty in Large Language Models](https://epub.jku.at/obvulihs/download/pdf/14266658)
- [Google Scholar](https://scholar.google.com/citations?user=W2mpUgIAAAAJ&hl=en)
- [GitHub](https://github.com/aichberger)

## Contact and corrections

- [Email Lukas Aichberger](mailto:lukas.aichberger@gmail.com): Corrections and material updates are welcome.