Amplify, Locate, Estimate (ALE) on Rare-Failure Rates of Fine-Tuned LLMs

The paper introduces Amplify, Locate, Estimate (ALE), a technique for efficiently estimating rare failure probabilities in safety‑fine‑tuned large language models. ALE amplifies failure rates by exploiting changes in token log‑probabilities, then uses a score‑function estimator and importance sampling to extrapolate back to the original rare rates. Experiments on 41 targets show ALE’s estimates closely match reference values, achieving accurate predictions for failure rates as low as 10⁻⁵ with far fewer labeled rollouts than direct sampling.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

• llms
• fine-tuning
• benchmarks
• safety