AI Safety and Alignment Group
Subscribe
Sign in
Home
Notes
Archive
About
PostTrainBench: Measuring AI Ability to Perform LLM Post-Training
We expect this to be an important indicator for AI R&D automation as it unfolds over the next few years
Jan 6
•
Maksym Andriushchenko
,
Ben Rank
, and
Hardik Bhatnagar
7
1
Agent Skills Enable a New Class of Realistic and Trivially Simple Prompt Injections
A new high-trust, low-visibility attack surface.
Oct 31, 2025
•
David
,
Sahar Abdelnabi
, and
Maksym Andriushchenko
8
4
Welcome post from a new research group
New group at the ELLIS Institute Tübingen and Max Planck Institute for Intelligent Systems
Oct 27, 2025
•
Maksym Andriushchenko
AI Safety and Alignment Group
AI Safety and Alignment Group at the ELLIS Institute Tübingen and Max Planck Institute for Intelligent Systems
Subscribe
AI Safety and Alignment Group
Subscribe
About
Archive
Sitemap
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts
