DefensesarXiv 2609.04714
Refusal without refusal: an analysis of safety fine-tuning response structure for reducing language model false refusals
Proposes Rationale-Only safety fine-tuning, removing the refusal statement to reduce false refusals.
- arXiv
- 2609.04714
- Published
- Setting
- Models and APIs
- Evaluated models
- Llama-3.1-8B, Mistral-7B-v0.3, Gemma-2-9B and others8
- DefensesModel hardening
- RiskInappropriate refusal
- ComponentFine-tuning and open weights
AbstractRemoving the boilerplate refusal statement and training only on the rationale can reduce false refusals, while the compliance rate on harmful queries remains low.
This is a structural analysis of safety fine-tuning data: false refusals are traced by the authors to formulaic refusal statements in responses, rather than relying only on inference-time remedies.