# U-CAN: Utility-Aware Contrastive Attenuation for Efficient Unlearning in Generative Recommendation

arXiv: https://arxiv.org/abs/2602.23400 · HTML: https://mlsanity.com/paper/2602.23400/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Membership inference attacks, one of 12 directions in LLM safety and security.
- Builds on 28 of its 55 references that are in the map.
- First posted on arXiv 26 Feb 2026.

## Placed in

- LLM safety and security / [Membership inference attacks](https://mlsanity.com/safety/membership-inference-attacks/)

## Builds on (most cited first, 28 of 28)

- [QLoRA: Efficient Finetuning of Quantized LLMs](https://mlsanity.com/paper/2305.14314/) (2023)
- [The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks](https://mlsanity.com/paper/1803.03635/) (2018)
- [Extracting Training Data from Large Language Models](https://mlsanity.com/paper/2012.07805/) (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)
- [Unifying Large Language Models and Knowledge Graphs: A Roadmap](https://mlsanity.com/paper/2306.08302/) (2023)
- [Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning](https://mlsanity.com/paper/2404.05868/) (2024)
- [Making AI Forget You: Data Deletion in Machine Learning](https://mlsanity.com/paper/1907.05012/) (2019)
- [Knowledge Unlearning for Mitigating Privacy Risks in Language Models](https://mlsanity.com/paper/2210.01504/) (2022)
- [Recommendation as Language Processing (RLP): A Unified Pretrain, Personalized Prompt & Predict Paradigm (P5)](https://mlsanity.com/paper/2203.13366/) (2022)
- [SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation](https://mlsanity.com/paper/2310.12508/) (2023)
- [TALLRec: An Effective and Efficient Tuning Framework to Align Large Language Model with Recommendation](https://mlsanity.com/paper/2305.00447/) (2023)
- [Descent-to-Delete: Gradient-Based Methods for Machine Unlearning](https://mlsanity.com/paper/2007.02923/) (2020)
- [MQuAKE: Assessing Knowledge Editing in Language Models via Multi-Hop Questions](https://mlsanity.com/paper/2305.14795/) (2023)
- [Quark: Controllable Text Generation with Reinforced Unlearning](https://mlsanity.com/paper/2205.13636/) (2022)
- [Unveiling Privacy Risks in LLM Agent Memory](https://mlsanity.com/paper/2502.13172/) (2025)
- [Trustworthy Graph Neural Networks: Aspects, Methods and Trends](https://mlsanity.com/paper/2205.07424/) (2022)
- [Large Language Models-guided Dynamic Adaptation for Temporal Knowledge Graph Reasoning](https://mlsanity.com/paper/2405.14170/) (2024)
- [Towards Robust and Parameter-Efficient Knowledge Unlearning for LLMs](https://mlsanity.com/paper/2408.06621/) (2024)
- [Making Recommender Systems Forget: Learning and Unlearning for Erasable Recommendation](https://mlsanity.com/paper/2203.11491/) (2022)
- [Retrieval-Augmented Conversational Recommendation with Prompt-based Semi-Structured Natural Language State Tracking](https://mlsanity.com/paper/2406.00033/) (2024)
- [Exact and Efficient Unlearning for Large Language Model-based Recommendation](https://mlsanity.com/paper/2404.10327/) (2024)
- [Dynamic Graph Unlearning: A General and Efficient Post-Processing Method via Gradient Transformation](https://mlsanity.com/paper/2405.14407/) (2024)
- [EARN: Efficient Inference Acceleration for LLM-based Generative Recommendation by Register Tokens](https://mlsanity.com/paper/2507.00715/) (2025)
- [NR4DER: Neural Re-ranking for Diversified Exercise Recommendation](https://mlsanity.com/paper/2506.06341/) (2025)
- [Beyond Whole Dialogue Modeling: Contextual Disentanglement for Conversational Recommendation](https://mlsanity.com/paper/2504.17427/) (2025)
- [Towards Distribution Matching between Collaborative and Language Spaces for Generative Recommendation](https://mlsanity.com/paper/2504.07363/) (2025)
- [Unlearning for Federated Online Learning to Rank: A Reproducibility Study](https://mlsanity.com/paper/2505.12791/) (2025)

## Built on by (most cited first, 1 of 1)

- [Calibration vs Decision Making: Revisiting the Reliability Paradox in Unlearned Language Models](https://mlsanity.com/paper/2605.20915/) (2026)

## Abstract (arXiv)

Generative Recommendation (GenRec) typically leverages Large Language Models (LLMs) to redefine personalization as an instruction-driven sequence generation task. However, fine-tuning on user logs inadvertently encodes sensitive attributes into model parameters, raising critical privacy concerns. Existing Machine Unlearning (MU) techniques struggle to navigate this tension due to the Polysemy Dilemma, where neurons superimpose sensitive data with general reasoning patterns, leading to catastrophic utility loss under traditional gradient or pruning methods. To address this, we propose Utility-aware Contrastive AttenuatioN (U-CAN), a precision unlearning framework that operates on low-rank adapters. U-CAN quantifies risk by contrasting activations and focuses on neurons with asymmetric responses that are highly sensitive to the forgetting set but suppressed on the retention set. To safeguard performance, we introduce a utility-aware calibration mechanism that combines weight magnitudes with retention-set activation norms, assigning higher utility scores to dimensions that contribute strongly to retention performance. Unlike binary pruning, which often fragments network structure, U-CAN develop adaptive soft attenuation with a differentiable decay function to selectively down-scale high-risk parameters on LoRA adapters, suppressing sensitive retrieval pathways and preserving the topological connectivity of reasoning circuits. Experiments on two public datasets across seven metrics demonstrate that U-CAN achieves strong privacy forgetting, utility retention, and computational efficiency.