Abstract:Fine-tuning aligned language models on benign tasks (e.g. math tutoring) systematically breaks safety guardrails, even when training data contains no harmful content. While mechanistic approaches have shed light on where alignment resides in model weights, they do not by provide a general formal framework for deriving guarantees about when fine-tuning degrades it -- leaving the field without principled tools for predicting or preventing alignment collapse. We develop a local geometric framework through geometric analysis of parameter-space trajectories and apply it to understand the fragility of alignment in fine-tuning. While first-order analysis suggests orthogonal updates are safe, we prove this is illusory: the curvature of the fine-tuning loss induces second-order acceleration that can induce second-order drift into alignment-sensitive regions. We formalize a construct of our framework as the Alignment Instability Condition (AIC), three geometric properties that, when present, are sufficient to guarantee degradation. Our main result proves quartic onset of alignment degradation along gradient-flow trajectories, determined by how sharply alignment depends on specific parameters and how strongly tasks couple to these parameters. These findings yield formal sufficient conditions under which static first-order protection can fail under gradient descent. We further empirically validate the framework's foundations, showing that the Fisher Information Matrix provides a proxy for the degree of safety degradation across diverse fine-tuning.
From: Bohdan Turbal [view email]
[v1]
Sun, 14 Jun 2026 01:18:53 UTC (618 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。