# Beyond Surface Alignment: Rebuilding LLMs Safety Mechanism via Probabilistically Ablating Refusal Direction

arXiv: https://arxiv.org/abs/2509.15202 · HTML: https://mlsanity.com/paper/2509.15202/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Placed in Safety alignment and jailbreak attacks, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 23 of its 37 references that are in the map.
- First posted on arXiv 18 Sep 2025.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Safety alignment and jailbreak attacks](https://mlsanity.com/alignment/safety-alignment-and-jailbreak-attacks/)
- LLM safety and security / [Jailbreaking](https://mlsanity.com/safety/jailbreaking/)

## Builds on (most cited first, 23 of 23)

- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Jailbroken: How Does LLM Safety Training Fail?](https://mlsanity.com/paper/2307.02483/) (2023)
- [HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal](https://mlsanity.com/paper/2402.04249/) (2024)
- [Jailbreaking Black Box Large Language Models in Twenty Queries](https://mlsanity.com/paper/2310.08419/) (2023)
- [Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!](https://mlsanity.com/paper/2310.03693/) (2023)
- [Steering Llama 2 via Contrastive Activation Addition](https://mlsanity.com/paper/2312.06681/) (2023)
- [Refusal in Language Models Is Mediated by a Single Direction](https://mlsanity.com/paper/2406.11717/) (2024)
- [Enhancing Chat Language Models by Scaling High-quality Instructional Conversations](https://mlsanity.com/paper/2305.14233/) (2023)
- [SmoothLLM: Defending Large Language Models Against Jailbreaking Attacks](https://mlsanity.com/paper/2310.03684/) (2023)
- [Detecting Language Model Attacks with Perplexity](https://mlsanity.com/paper/2308.14132/) (2023)
- [Improving Alignment and Robustness with Circuit Breakers](https://mlsanity.com/paper/2406.04313/) (2024)
- [SafeDecoding: Defending against Jailbreak Attacks via Safety-Aware Decoding](https://mlsanity.com/paper/2402.08983/) (2024)
- [Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs](https://mlsanity.com/paper/2407.15549/) (2024)
- [Tamper-Resistant Safeguards for Open-Weight LLMs](https://mlsanity.com/paper/2408.00761/) (2024)
- [The Geometry of Refusal in Large Language Models: Concept Cones and Representational Independence](https://mlsanity.com/paper/2502.17420/) (2025)
- [Efficient Adversarial Training in LLMs with Continuous Attacks](https://mlsanity.com/paper/2405.15589/) (2024)
- [How Alignment and Jailbreak Work: Explain LLM Safety through Intermediate Hidden States](https://mlsanity.com/paper/2406.05644/) (2024)
- [Lisa: Lazy Safety Alignment for Large Language Models against Harmful Fine-tuning Attack](https://mlsanity.com/paper/2405.18641/) (2024)
- [Uncovering Safety Risks of Large Language Models through Concept Activation Vector](https://mlsanity.com/paper/2404.12038/) (2024)
- [Robust LLM safeguarding via refusal feature adversarial training](https://mlsanity.com/paper/2409.20089/) (2024)
- [Bypassing the Safety Training of Open-Source LLMs with Priming Attacks](https://mlsanity.com/paper/2312.12321/) (2023)
- [Steering Without Side Effects: Improving Post-Deployment Control of Language Models](https://mlsanity.com/paper/2406.15518/) (2024)

## Built on by (most cited first, 8 of 8)

- [Read the Scene, Not the Script: Outcome-Aware Safety for LLMs](https://mlsanity.com/paper/2510.04320/) (2025)
- [RouteHijack: Routing-Aware Attack on Mixture-of-Experts LLMs](https://mlsanity.com/paper/2605.02946/) (2026)
- [Fail-Closed Alignment for Large Language Models](https://mlsanity.com/paper/2602.16977/) (2026)
- [Dynamic Adversarial Fine-Tuning Reorganizes Refusal Geometry](https://mlsanity.com/paper/2604.27019/) (2026)
- [RefusalGuard: Geometry-Preserving Fine-Tuning for Safety in LLMs](https://mlsanity.com/paper/2605.01913/) (2026)
- [Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks](https://mlsanity.com/paper/2607.23015/) (2026)
- [Making Your LLMs More Objective: Stabilizing LLM Safety Behavior Across Traits with Trait-Invariant Safety Tuning](https://mlsanity.com/paper/2608.11705/) (2026)
- [Fool's Gold: Defensive Deception Against Safety-Removal Attacks on Open-Weight Models](https://mlsanity.com/paper/2608.17202/) (2026)

## Abstract (arXiv)

Jailbreak attacks pose persistent threats to large language models (LLMs). Current safety alignment methods have attempted to address these issues, but they experience two significant limitations: insufficient safety alignment depth and unrobust internal defense mechanisms. These limitations make them vulnerable to adversarial attacks such as prefilling and refusal direction manipulation. We introduce DeepRefusal, a robust safety alignment framework that overcomes these issues. DeepRefusal forces the model to dynamically rebuild its refusal mechanisms from jailbreak states. This is achieved by probabilistically ablating the refusal direction across layers and token depths during fine-tuning. Our method not only defends against prefilling and refusal direction attacks but also demonstrates strong resilience against other unseen jailbreak strategies. Extensive evaluations on four open-source LLM families and six representative attacks show that DeepRefusal reduces attack success rates by approximately 95%, while maintaining model capabilities with minimal performance degradation.