居中消融:為何末層方向消融效果不足或導致崩潰,以及一種封閉形式的修正方法
Centered Ablation: Why Last-Layer Directional Ablation Under-Fires or Collapses, and a Closed-Form Fix
研究提出 Centered Ablation(CA),以封閉形式從消融方向中投影去除共同模式,解釋並修正標準末層方向消融效果不足或令文本崩潰的問題。在 8 個模型上,CA 在標準方法效果不足時將攻擊成功率提高 29 和 46 個百分點,並將 Gemma-2-9B 的崩潰率由 50% 降至 1%。研究亦指出,兩種標準安全評審對崩潰文本的評分方向相反,可能造成 46 個百分點的效果差異。
Published on Nov 15
Authors:
Abstract
Directional ablation removes a hidden state's component along a "behaviour direction" while a language model runs. It is the standard way to switch a behaviour such as refusal off, but it is erratic: on some models it barely fires, on others the text collapses into repetition. We show why, and fix it. At the last residual stream, ablation is exactly a logit bias: a fixed vocabulary footprint scaled by a gate, the cosine between the hidden state and the direction. The gate is measured from the origin, but last-layer states share a large common mode far from it, so a direction that overlaps the common mode fires on benign text and under-fires on its targets. Centered Ablation (CA) projects the common mode out of the direction, in closed form. Across 8 models, CA raises attack success by 29 and 46 points where the standard recipe under-fires, cuts collapse on Gemma-2-9B from 50\% to 1\%, and preserves 12 to 40 points more benign answers on high-overlap models, 39 to 46 more than the affine offset of prior work. Two numbers computed before generation predict, on models the analysis had not seen, which gain faithfulness and which gain attack success, and a collapse index marks when text degenerates. We also show that two standard safety judges score collapsed text in opposite directions, which can fabricate a 46-point effect.
View arXiv page View PDF Project page GitHub 0 Add to collection
Get this paper in your agent:
hf papers read 2611.20148
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash
Models citing this paper 0
No model linking this paper
Cite arxiv.org/abs/2611.20148 in a model README.md to link it from this page.
Datasets citing this paper 0
No dataset linking this paper
Cite arxiv.org/abs/2611.20148 in a dataset README.md to link it from this page.
Spaces citing this paper 0
No Space linking this paper
Cite arxiv.org/abs/2611.20148 in a Space README.md to link it from this page.
Collections including this paper 0
No Collection including this paper
Add this paper to a collection to link it from this page.
來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co