# Causal Routing for Unlearning

arXiv: https://arxiv.org/abs/2609.34475 · HTML: https://mlsanity.com/paper/2609.34475/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Machine unlearning, one of 12 directions in LLM safety and security.
- Builds on 36 of its 46 references that are in the map.
- First posted on arXiv 28 Sep 2026.

## Placed in

- LLM safety and security / [Machine unlearning](https://mlsanity.com/safety/machine-unlearning/)

## Builds on (most cited first, 30 of 36)

- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Parameter-Efficient Transfer Learning for NLP](https://mlsanity.com/paper/1902.00751/) (2019)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Locating and Editing Factual Associations in GPT](https://mlsanity.com/paper/2202.05262/) (2022)
- [GLU Variants Improve Transformer](https://mlsanity.com/paper/2002.05202/) (2020)
- [Transformer Feed-Forward Layers Are Key-Value Memories](https://mlsanity.com/paper/2012.14913/) (2020)
- [Machine Unlearning](https://mlsanity.com/paper/1912.03817/) (2019)
- [BERT Rediscovers the Classical NLP Pipeline](https://mlsanity.com/paper/1905.05950/) (2019)
- [Steering Language Models With Activation Engineering](https://mlsanity.com/paper/2308.10248/) (2023)
- [Language Modeling with Gated Convolutional Networks](https://mlsanity.com/paper/1612.08083/) (2016)
- [Knowledge Neurons in Pretrained Transformers](https://mlsanity.com/paper/2104.08696/) (2021)
- [Certified Data Removal from Machine Learning Models](https://mlsanity.com/paper/1911.03030/) (2019)
- [TOFU: A Task of Fictitious Unlearning for LLMs](https://mlsanity.com/paper/2401.06121/) (2024)
- [Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning](https://mlsanity.com/paper/2404.05868/) (2024)
- [SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation](https://mlsanity.com/paper/2310.12508/) (2023)
- [DoLa: Decoding by Contrasting Layers Improves Factuality in Large Language Models](https://mlsanity.com/paper/2309.03883/) (2023)
- [Remember What You Want to Forget: Algorithms for Machine Unlearning](https://mlsanity.com/paper/2103.03279/) (2021)
- [Large Language Model Unlearning](https://mlsanity.com/paper/2310.10683/) (2023)
- [MUSE: Machine Unlearning Six-Way Evaluation for Language Models](https://mlsanity.com/paper/2407.06460/) (2024)
- [Rethinking Machine Unlearning for Large Language Models](https://mlsanity.com/paper/2402.08787/) (2024)
- [In-Context Unlearning: Language Models as Few Shot Unlearners](https://mlsanity.com/paper/2310.07579/) (2023)
- [Programming Refusal with Conditional Activation Steering](https://mlsanity.com/paper/2409.05907/) (2024)
- [Simplicity Prevails: Rethinking Negative Preference Optimization for LLM Unlearning](https://mlsanity.com/paper/2410.07163/) (2024)
- [On the Necessity of Auditable Algorithmic Definitions for Machine Unlearning](https://mlsanity.com/paper/2110.11891/) (2021)
- [All Bark and No Bite: Rogue Dimensions in Transformer Language Models Obscure Representational Quality](https://mlsanity.com/paper/2109.04404/) (2021)
- [Continual Learning and Private Unlearning](https://mlsanity.com/paper/2203.12817/) (2022)
- [Language-Specific Neurons: The Key to Multilingual Capabilities in Large Language Models](https://mlsanity.com/paper/2402.16438/) (2024)
- [Finding Skill Neurons in Pre-trained Transformer-based Language Models](https://mlsanity.com/paper/2211.07349/) (2022)
- [LLM Unlearning via Neural Activation Redirection](https://mlsanity.com/paper/2502.07218/) (2025)
- [Neuron-Level Knowledge Attribution in Large Language Models](https://mlsanity.com/paper/2312.12141/) (2023)

## Abstract (arXiv)

LLMs cannot forget the way we delete a file. Strangely, we are asked to remove something that was never put anywhere in particular. What the model took from a piece of text is now smeared across billions of weights. Existing methods rewrite all of them to change one thing, and none of them say which part produced that change. To address this, we introduce Causal Routing for Unlearning (CRU) by asking where the concept is expressed in the model and suppressing only that part. One untrained forward pass over the forget set ranks neurons by how their activations vary. Then, small routing modules on those neurons gate and suppress only the concepts that need to be forgotten. In CRU, the base model is frozen, and any change in behavior is caused only by the gated neurons; hence, why the routing is causal. Due to our parameter efficiency (only ~0.01% as many parameters as the base model), unlearning a concept costs 14 GiB, whereas the baselines require 71 GiB. On TOFU, CRU is indistinguishable from the retained model (p > 0.05, KS test) and is never Pareto-dominated, whereas every compared baseline matches its forgetting on the larger-forget batches only by collapsing utility. On RWKU, it achieves an adversarial-probe recall of 0.052, compared to 0.250 for the strongest baseline, meaning the knowledge is gone, not merely harder to reach. Thus, deciding on the intervention at query time, rather than fixing it beforehand, is the axis along which we argue that unlearning should proceed.