# Knowledge Vector Weakening: Efficient Training-free Unlearning for Large Vision-Language Models

arXiv: https://arxiv.org/abs/2601.21794 · HTML: https://mlsanity.com/paper/2601.21794/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in MLLM jailbreaks and adversarial robustness, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 30 of its 42 references that are in the map.
- First posted on arXiv 29 Jan 2026.

## Placed in

- Multimodal LLMs / vision-language models / [MLLM jailbreaks and adversarial robustness](https://mlsanity.com/mllm/mllm-jailbreaks-and-adversarial-robustness/)
- LLM safety and security / [Machine unlearning](https://mlsanity.com/safety/machine-unlearning/)

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

- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Locating and Editing Factual Associations in GPT](https://mlsanity.com/paper/2202.05262/) (2022)
- [Transformer Feed-Forward Layers Are Key-Value Memories](https://mlsanity.com/paper/2012.14913/) (2020)
- [Editing Models with Task Arithmetic](https://mlsanity.com/paper/2212.04089/) (2022)
- [Knowledge Neurons in Pretrained Transformers](https://mlsanity.com/paper/2104.08696/) (2021)
- [TOFU: A Task of Fictitious Unlearning for LLMs](https://mlsanity.com/paper/2401.06121/) (2024)
- [Transformer Feed-Forward Layers Build Predictions by Promoting Concepts in the Vocabulary Space](https://mlsanity.com/paper/2203.14680/) (2022)
- [Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning](https://mlsanity.com/paper/2404.05868/) (2024)
- [A Survey of Machine Unlearning](https://mlsanity.com/paper/2209.02299/) (2022)
- [Unrolling SGD: Understanding Factors Influencing Machine Unlearning](https://mlsanity.com/paper/2109.13398/) (2021)
- [What Does it Mean for a Language Model to Preserve Privacy?](https://mlsanity.com/paper/2202.05520/) (2022)
- [Continual Learning and Private Unlearning](https://mlsanity.com/paper/2203.12817/) (2022)
- [Variational Bayesian Unlearning](https://mlsanity.com/paper/2010.12883/) (2020)
- [xGen-MM (BLIP-3): A Family of Open Large Multimodal Models](https://mlsanity.com/paper/2408.08872/) (2024)
- [LoRA vs Full Fine-tuning: An Illusion of Equivalence](https://mlsanity.com/paper/2410.21228/) (2024)
- [Calibrating Factual Knowledge in Pretrained Language Models](https://mlsanity.com/paper/2210.03329/) (2022)
- [Protecting Privacy in Multimodal Large Language Models with MLLMU-Bench](https://mlsanity.com/paper/2410.22108/) (2024)
- [Are we making progress in unlearning? Findings from the first NeurIPS unlearning competition](https://mlsanity.com/paper/2406.09073/) (2024)
- [Q-VLM: Post-training Quantization for Large Vision-Language Models](https://mlsanity.com/paper/2410.08119/) (2024)
- [Towards Robust and Parameter-Efficient Knowledge Unlearning for LLMs](https://mlsanity.com/paper/2408.06621/) (2024)
- [Benchmarking Vision Language Model Unlearning via Fictitious Facial Identity Dataset](https://mlsanity.com/paper/2411.03554/) (2024)
- [Modality-Aware Neuron Pruning for Unlearning in Multimodal Large Language Models](https://mlsanity.com/paper/2502.15910/) (2025)
- [Breaking the Trilemma of Privacy, Utility, Efficiency via Controllable Machine Unlearning](https://mlsanity.com/paper/2310.18574/) (2023)
- [Knowledge Entropy Decay during Language Model Pretraining Hinders New Knowledge Acquisition](https://mlsanity.com/paper/2410.01380/) (2024)
- [Hyperparameters in Continual Learning: A Reality Check](https://mlsanity.com/paper/2403.09066/) (2024)
- [PEBench: A Fictitious Dataset to Benchmark Machine Unlearning for Multimodal Large Language Models](https://mlsanity.com/paper/2503.12545/) (2025)
- [Improving Fisher Information Estimation and Efficiency for LoRA-based LLM Unlearning](https://mlsanity.com/paper/2508.21300/) (2025)
- [Reference-Specific Unlearning Metrics Can Hide the Truth: A Reality Check](https://mlsanity.com/paper/2510.12981/) (2025)

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

- [Suppression Is Not Forgetting: Residual Recoverability in Visual Concept Unlearning for VLMs](https://mlsanity.com/paper/2604.03114/) (2026)
- [SIEVE: Selective attention-value Suppression for Vision-Language Models Unlearning](https://mlsanity.com/paper/2610.01962/) (2026)
- [Open-MMUnlearning: Unifying Methods and Evaluation for MLLM Unlearning](https://mlsanity.com/paper/2610.10358/) (2026)

## Abstract (arXiv)

Large Vision-Language Models (LVLMs) are widely adopted for their strong multimodal capabilities, yet they raise serious concerns such as privacy leakage and harmful content generation. Machine unlearning has emerged as a promising solution for removing the influence of specific data from trained models. However, existing approaches largely rely on gradient-based optimization, incurring substantial computational costs for large-scale LVLMs. To address this limitation, we propose Knowledge Vector Weakening (KVW), a training-free unlearning method that directly intervenes in the full model without gradient computation. KVW identifies knowledge vectors that are activated during the model's output generation on the forget set and progressively weakens their contributions, thereby preventing the model from exploiting undesirable knowledge. Experiments on the MLLMU and CLEAR benchmarks demonstrate that KVW achieves a stable forget-retain trade-off while significantly improving computational efficiency over gradient-based and LoRA-based unlearning methods.