SafeTuneBed is a plugin-based benchmark and toolkit that standardizes the data, defenses, and metrics used to evaluate how well LLM fine-tuning preserves safety alignment.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
SafeTuneBed: A Toolkit for Benchmarking LLM Safety Alignment in Fine-Tuning
SafeTuneBed is a plugin-based benchmark and toolkit that standardizes the data, defenses, and metrics used to evaluate how well LLM fine-tuning preserves safety alignment.