LLMs exhibit misfired alignment on stereotype questions at 4.7-18.9% rates on the new VETO benchmark of 2,032 contrastive pairs, unlike humans at 0%, due to overgeneralized safety cues after instruction tuning.
Felkner, Ho-Chun Herbert Chang, Eugene Jang, and Jonathan May
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
verdicts
UNVERDICTED 3roles
background 1polarities
background 1representative citing papers
One-shot GRPO on a single biased example induces generalizing stereotype bias in post-trained LLMs, with susceptibility varying by initial bias likelihood.
The paper introduces a taxonomy of AI safety for LLMs organized into Trustworthy AI, Responsible AI, and Safe AI perspectives, accompanied by a review of state-of-the-art methods, challenges, and future directions.
citing papers explorer
-
The Wrong Kind of Right: Quantifying and Localizing Misfired Alignment in LLMs
LLMs exhibit misfired alignment on stereotype questions at 4.7-18.9% rates on the new VETO benchmark of 2,032 contrastive pairs, unlike humans at 0%, due to overgeneralized safety cues after instruction tuning.
-
It Takes One to Bias Them All: Breaking Bad with One-Shot GRPO
One-shot GRPO on a single biased example induces generalizing stereotype bias in post-trained LLMs, with susceptibility varying by initial bias likelihood.
-
AI Safety Landscape for Large Language Models: Taxonomy, State-of-the-art, and Future Directions
The paper introduces a taxonomy of AI safety for LLMs organized into Trustworthy AI, Responsible AI, and Safe AI perspectives, accompanied by a review of state-of-the-art methods, challenges, and future directions.