BioBlue: Systematic runaway-optimiser-like LLM failure modes on biologically and economically aligned AI safety benchmarks for LLMs with simplified observation format

Zenodo (CERN European Organization for Nuclear Research) 2026 Open access

Read full text
Authors: Roland Pihlakas, Sruthi Susan Kuriakose
DOI: 10.5281/zenodo.18121362
Cited by: 0

Abstract

Updates to Claude support (including caching), removing llama-tokens package dependency for time being. Minor updates to LLMUtilities.py and Utilities.py Updating citation.cff and licence notices with latest preprint date.

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
4 July 2026