mlsanityData to 7 Oct 2026

DualOptim+: Bridging Shared and Decoupled Optimizer States for Better Machine Unlearning in Large Language Models

arXiv 2605.21539 ↗First posted 20 May 2026cs.LGcited by 0

On the map: LLM safety and security → Machine unlearning

Abstract (arXiv)

We propose DualOptim+, a novel optimization framework for improving machine unlearning in large language models. It introduces a base state to capture common representations shared by forgetting and retaining objectives and delta states to preserve objective-specific residuals. This architecture allows the optimizer to adaptively bridge shared and decoupled states based on the directional conflict between forgetting and retaining gradients. We further introduce DualOptim+ 8bit, a quantized variant that reduces memory overhead without compromising performance. Extensive experiments across fictitious and real-world unlearning, safety alignment, and multi-task learning tasks demonstrate that DualOptim+ consistently achieves a superior trade-off between different objectives. Codes are available at https://github.com/CityU-MLO/DualOptimPlus.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Machine unlearning, one of 12 directions in LLM safety and security. Builds on 43 of its 49 references that are in the map. First posted on arXiv 20 May 2026.

Builds on

43 of 49 references.

Closest ancestors · 19

Same direction, newest first

  1. 2025DRAGON: Guard LLM Unlearning in Context via Negative Detection and Reasoning8 Nov
  2. 2025BLUR: A Bi-Level Optimization Approach for LLM Unlearning9 Jun
  3. 2025SEPS: A Separability Measure for Robust Unlearning in LLMs20 May
  4. 2025GUARD: Generation-time LLM Unlearning via Adaptive Restriction and Detection19 May
  5. 2025DualOptim: Enhancing Efficacy and Stability in Machine Unlearning with Dual Optimizers22 Apr
  6. 2024A Closer Look at Machine Unlearning for Large Language Models10 Oct
  7. 2024Simplicity Prevails: Rethinking Negative Preference Optimization for LLM Unlearning9 Oct
  8. 2024Erasing Conceptual Knowledge from Language Models3 Oct
  9. 2024Unified Gradient-Based Machine Unlearning with Remain Geometry Enhancement29 Sep
  10. 2024Learning to Refuse: Towards Mitigating Privacy Risks in LLMs14 Jul
  11. 2024Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning8 Apr
  12. 2024TOFU: A Task of Fictitious Unlearning for LLMs11 Jan
7 more
  1. 2023SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation19 Oct
  2. 2023Large Language Model Unlearning14 Oct
  3. 2023In-Context Unlearning: Language Models as Few Shot Unlearners11 Oct
  4. 2023Model Sparsity Can Simplify Machine Unlearning11 Apr
  5. 2023Towards Unbounded Machine Unlearning20 Feb
  6. 2021Fast Yet Effective Machine Unlearning17 Nov
  7. 2021On the Necessity of Auditable Algorithmic Definitions for Machine Unlearning22 Oct

Other references · 24

  1. 2024The Llama 3 Herd of Modelscited by 14,192
  2. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  3. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  4. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,916
  5. 2020Measuring Massive Multitask Language Understandingcited by 7,768
  6. 2018Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challengecited by 4,710
  7. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,634
  8. 2021TruthfulQA: Measuring How Models Mimic Human Falsehoodscited by 2,919
  9. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,895
  10. 2022Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answeringcited by 1,881
  11. 2019Machine Unlearningcited by 1,020
  12. 2021CodeXGLUE: A Machine Learning Benchmark Dataset for Code Understanding and Generationcited by 732
12 more
  1. 2023Textbooks Are All You Need II: phi-1.5 technical reportcited by 559
  2. 2024The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearningcited by 488
  3. 2023XSTest: A Test Suite for Identifying Exaggerated Safety Behaviours in Large Language Modelscited by 477
  4. 2023Safety-Tuned LLaMAs: Lessons From Improving the Safety of Large Language Models that Follow Instructionscited by 377
  5. 2024Improving Alignment and Robustness with Circuit Breakerscited by 302
  6. 2015Federated Optimization:Distributed Optimization Beyond the Datacentercited by 193
  7. 2023Selective Amnesia: A Continual Learning Approach to Forgetting in Deep Generative Modelscited by 175
  8. 2020Mime: Mimicking Centralized Stochastic Algorithms in Federated Learningcited by 152
  9. 2022NumGLUE: A Suite of Fundamental yet Challenging Mathematical Reasoning Taskscited by 110
  10. 2021Human-in-the-Loop for Data Collection: a Multi-Target Counter Narrative Dataset to Fight Online Hate Speechcited by 81
  11. 2023TRACE: A Comprehensive Benchmark for Continual Learning in Large Language Modelscited by 59
  12. 2024Convergence of Distributed Adaptive Optimization with Local Updatescited by 10

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024A Probabilistic Perspective on Unlearning and Alignment for Large Language Models
  2. 2024Alternate Preference Optimization for Unlearning Factual Knowledge in Large Language Models
  3. 2025Towards Reasoning-Preserving Unlearning in Multimodal Large Language Models
  4. 2024Rethinking Machine Unlearning for Large Language Models
  5. 2024Towards Robust and Parameter-Efficient Knowledge Unlearning for LLMs
  6. 2025LLM Unlearning Reveals a Stronger-Than-Expected Coreset Effect in Current Benchmarks
  7. 2026Exploring and Bridging Knowledge Holes in Unlearned Multimodal Large Language Models
  8. 2025Sparse-Autoencoder-Guided Internal Representation Unlearning for Large Language Models