Fail-Closed Alignment for Large Language Models
We identify a structural weakness in current large language model (LLM) alignment: modern refusal mechanisms are fail-open. While existing approaches encode refusal behaviors across multiple latent features, suppressing a single dominant feature$-$via prompt-based jailbreaks$-$can cause alignment to collapse, leading to unsafe generation. Motivated by this, we propose fail-closed alignment as a design principle for robust LLM safety: refusal mechanisms should remain effective even under partial failures via redundant, independent causal pathways. We present a concrete instantiation of this principle: a progressive alignment framework that iteratively identifies and ablates previously lear...
| Authors | Zachary Coalson, Beth Sohler, Aiden Gabriel et al. |
|---|---|
| Primary category | cs.LG |
| Published | 19 February 2026 |
| Last updated at source | 19 February 2026 |
Models and datasets referencing this paper
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.