Controlled audit of language-tuned LLMs reveals a Fine-Tuning Paradox where Ukrainian-oriented models resist Russian disinformation less in Russian than Russian-oriented models do, indicating corpus and prompt factors outweigh cultural provenance.
Uncensored AI in the Wild: Tracking Publicly Available and Locally Deployable LLMs
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
An empirical red-teaming study measures political Overton Windows across more than 30 open-source LLMs from 10 families and finds left-leaning bias, inverse size correlation, regional variation, and variable jailbreak effectiveness.
citing papers explorer
-
Friend or Foe? Language as an ideological switch in open-weight LLMs under Russian disinformation stress
Controlled audit of language-tuned LLMs reveals a Fine-Tuning Paradox where Ukrainian-oriented models resist Russian disinformation less in Russian than Russian-oriented models do, indicating corpus and prompt factors outweigh cultural provenance.
-
How Far Will They Go? Red-Teaming Online Influence with Large Language Models
An empirical red-teaming study measures political Overton Windows across more than 30 open-source LLMs from 10 families and finds left-leaning bias, inverse size correlation, regional variation, and variable jailbreak effectiveness.