# BLOCK-EM: Preventing Emergent Misalignment via Latent Blocking

arXiv: https://arxiv.org/abs/2602.00767 · HTML: https://mlsanity.com/paper/2602.00767/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 33 references that are in the map.
- First posted on arXiv 31 Jan 2026.

## Builds on (most cited first, 18 of 18)

- [Locating and Editing Factual Associations in GPT](https://mlsanity.com/paper/2202.05262/) (2022)
- [Emergent Abilities of Large Language Models](https://mlsanity.com/paper/2206.07682/) (2022)
- [Sparse Autoencoders Find Highly Interpretable Features in Language Models](https://mlsanity.com/paper/2309.08600/) (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)
- [Toy Models of Superposition](https://mlsanity.com/paper/2209.10652/) (2022)
- [Steering Language Models With Activation Engineering](https://mlsanity.com/paper/2308.10248/) (2023)
- [Gemma Scope: Open Sparse Autoencoders Everywhere All At Once on Gemma 2](https://mlsanity.com/paper/2408.05147/) (2024)
- [Towards Best Practices of Activation Patching in Language Models: Metrics and Methods](https://mlsanity.com/paper/2309.16042/) (2023)
- [Persona Vectors: Monitoring and Controlling Character Traits in Language Models](https://mlsanity.com/paper/2507.21509/) (2025)
- [Layer by Layer: Uncovering Hidden Representations in Language Models](https://mlsanity.com/paper/2502.02013/) (2025)
- [How to use and interpret activation patching](https://mlsanity.com/paper/2404.15255/) (2024)
- [Persona Features Control Emergent Misalignment](https://mlsanity.com/paper/2506.19823/) (2025)
- [Thought Crime: Backdoors and Emergent Misalignment in Reasoning Models](https://mlsanity.com/paper/2506.13206/) (2025)
- [Inoculation Prompting: Instructing LLMs to misbehave at train-time improves test-time alignment](https://mlsanity.com/paper/2510.05024/) (2025)
- [Steering Out-of-Distribution Generalization with Concept Ablation Fine-Tuning](https://mlsanity.com/paper/2507.16795/) (2025)
- [Interpretable LLM Guardrails via Sparse Representation Steering](https://mlsanity.com/paper/2503.16851/) (2025)
- [The Devil in the Details: Emergent Misalignment, Format and Coherence in Open-Weights LLMs](https://mlsanity.com/paper/2511.20104/) (2025)

## Built on by (most cited first, 3 of 3)

- [Emergent Misalignment Recruits a Pre-existing Persona Subspace](https://mlsanity.com/paper/2607.21356/) (2026)
- [Actionable Activation Directions for Detecting and Mitigating Emergent Misalignment Across Language Model Families](https://mlsanity.com/paper/2606.20225/) (2026)
- [Subliminal Learning as Trait-Direction Drift: A Mechanism and Targeted Control under SFT Distillation](https://mlsanity.com/paper/2609.01091/) (2026)

## Abstract (arXiv)

Emergent misalignment can arise when a language model is fine-tuned on a narrowly scoped supervised objective: the model learns the target behavior, yet also develops undesirable out-of-domain behaviors. We investigate a mechanistic approach to preventing emergent misalignment by identifying a small set of internal features that reliably control the misaligned behavior and then discouraging the model from strengthening these features during fine-tuning. Across six fine-tuning domains, blocking (i.e., constraining) a fixed set of features achieves up to 95% relative reduction in emergent misalignment with no degradation in model quality or target-task performance. We strengthen validity with disjoint selection/evaluation splits, multiple independent judges, multiple random seeds for key settings, quality metrics, and extensive ablations demonstrating that the reduction in misalignment is specific to the identified mechanism. We also characterize a limiting regime in which misalignment re-emerges under prolonged fine-tuning, present evidence consistent with rerouting through alternative features or layers, and evaluate modifications that partially restore the misalignment-blocking effect. Overall, our results show that targeted training-time constraints on internal mechanisms can mitigate emergent misalignment without degrading target-task performance.