Paper Detail
Yein Park, Jungwoo Park, Jaewoo Kang
Large language models (LLMs), despite being safety-aligned, exhibit brittle refusal behaviors that can be circumvented by simple linguistic changes. As tense jailbreaking demonstrates that models refusing harmful requests often comply when rephrased in past tense, a critical generalization gap is revealed in current alignment methods whose underlying mechanisms are poorly understood. In this work, we introduce Activation-Scaling Guard (ASGuard), an insightful, mechanistically-informed framework that surgically mitigates this specific vulnerability. In the first step, we use circuit analysis to identify the specific attention heads causally linked to the targeted jailbreaking such as a tense-changing attack. Second, we train a precise, channel-wise scaling vector to recalibrate the activation of tense vulnerable heads. Lastly, we apply it into a "preventative fine-tuning", forcing the model to learn a more robust refusal mechanism. Across four LLMs, ASGuard effectively reduces the attack success rate of targeted jailbreaking while preserving general capabilities and minimizing over refusal, achieving a Pareto-optimal balance between safety and utility. Our findings underscore how adversarial suffixes suppress the propagation of the refusal-mediating direction, based on mechanistic analysis. Furthermore, our work showcases how a deep understanding of model internals can be leveraged to develop practical, efficient, and targeted methods for adjusting model behavior, charting a course for more reliable and interpretable AI safety.
No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.
No ranking explanation is available yet.
No tags.
@misc{park2026asguard,
title = {ASGuard: Activation-Scaling Guard to Mitigate Targeted Jailbreaking Attack},
author = {Yein Park and Jungwoo Park and Jaewoo Kang},
year = {2026},
abstract = {Large language models (LLMs), despite being safety-aligned, exhibit brittle refusal behaviors that can be circumvented by simple linguistic changes. As tense jailbreaking demonstrates that models refusing harmful requests often comply when rephrased in past tense, a critical generalization gap is revealed in current alignment methods whose underlying mechanisms are poorly understood. In this work, we introduce Activation-Scaling Guard (ASGuard), an insightful, mechanistically-informed framework },
url = {https://huggingface.co/papers/2509.25843},
keywords = {large language models, jailbreaking, attention heads, circuit analysis, activation scaling, preventative fine-tuning, refusal behavior, adversarial suffixes, model internals, safety alignment, code available, huggingface daily},
eprint = {2509.25843},
archiveprefix = {arXiv},
}
{}