Machine Unlearning for LLMs
Research on methods to make LLMs forget specific knowledge or information to enhance privacy, security, and address copyright concerns
Machine Unlearning for LLMs
Research on Large Language Models in Machine Unlearning for LLMs
Strategic Model Forgetting for LLM Security
How Manipulating Internal Representations Makes LLMs Forget
Selective Forgetting for Safer LLMs
A robust approach to removing sensitive knowledge from language models
The Unlearning Illusion in AI Safety
Why removing hazardous knowledge from LLMs may be harder than we think
Rethinking LLM Unlearning Evaluations
Current benchmarks overestimate unlearning effectiveness
Beyond Greedy Decoding: A Probabilistic Approach to LLMs
Improving security evaluation for unlearning and alignment
Simplifying LLM Unlearning
A more effective approach to removing unwanted content from AI models
Erasing the Unwanted in LLMs
Machine Unlearning as a Solution for Data Privacy and Legal Compliance
Unmasking LLM Memory Erasure
Evaluating if harmful information is truly removed from language models
The Illusion of Forgetting in LLMs
How quantization can resurrect 'unlearned' knowledge in language models
Strategic Unlearning in LLMs
Precision-targeted weight modification for secure AI models
The Efficacy of LLM Unlearning
A critical evaluation of techniques to remove harmful information from AI models
Strengthening LLM Unlearning Security
Making Representation Misdirection methods robust against backdoor-like vulnerabilities
Selective Forgetting in AI
A Framework for Agentic LLM Unlearning
Forgetting What Tools Know
A Novel Framework for Security-Focused Tool Unlearning in LLMs
Safer AI Through Selective Forgetting
Precision-targeted knowledge removal in large language models
Preventing Copyright Violations in LLMs
A lightweight solution to disrupt memorized content generation
Selective Forgetting in Language Models
A novel approach to removing private information from LLMs
Selective Forgetting in MLLMs
A Novel Approach to Multimodal Machine Unlearning
Smarter Forgetting for AI Models
A novel approach to targeted unlearning in LLMs without sacrificing performance
Optimizing LLM Unlearning: The Retain Set Perspective
Strategic data retention for effective entity unlearning in LLMs
SafeEraser: Making AI Forget Harmful Content
Advancing safety through multimodal machine unlearning
Protecting IP in the Age of LLMs
Selective memory suppression without compromising performance
Balanced Data Forgetting in LLMs
Selective unlearning without compromising model utility
Selective Forgetting for LLMs
A benchmark for removing specific content from AI models
The Illusion of Forgetting in LLMs
Why soft token attacks fail as reliable auditing tools for machine unlearning
Securing MLLMs Through Smart Forgetting
Novel neuron pruning technique for targeted information removal in multimodal AI
Enhancing LLM Unlearning
A framework to remove sensitive data while preserving model utility
Advancing LLM Security through Better Unlearning
A comprehensive framework for auditing knowledge removal in large language models
Surgical Knowledge Removal in LLMs
A gradient-based approach to selective unlearning without compromising model integrity
Securing Knowledge Erasure in LLMs
Beyond Surface Deletion: Comprehensive Unlearning for True Knowledge Forgetting
Erasing Sensitive Data from AI Models
A More Effective Approach to AI Unlearning
Forgetting by Design: LLM Unlearning
Selectively removing sensitive data without full retraining
Selective Memory: Unlearning Sensitive Content in LLMs
Parameter-Efficient Techniques for Enhanced AI Privacy
Enhanced LLM Unlearning for Security
Beyond Forgetting: Removing Related Knowledge for Complete Unlearning
Selective Unlearning in LLMs
Efficiently Removing Sensitive Data Without Full Retraining
Balancing Unlearning & Retention in LLMs
A Gradient-Based Approach to Selective Knowledge Removal
Safer AI: Selective Memory Control for LLMs
Targeted knowledge removal without compromising overall performance
Securing MLLMs Through Machine Unlearning
A novel benchmark to evaluate privacy protection in multimodal AI
Selective Concept Unlearning for AI Security
Fine-grained knowledge removal in vision-language models using sparse autoencoders
Privacy-Preserving AI: Making Models Forget
A novel contrastive unlearning framework for language models
Selective Knowledge Erasure in LLMs
Advancing Security Through Model Merging Techniques
Selective Skill Unlearning in LLMs
Training-free techniques to control model capabilities
Selective Forgetting in AI Art
Teaching GANs to unlearn problematic content without retraining
Erasing Sensitive Data from LLMs
A Systematic Approach to Making AI Forget
Selective Forgetting in AI
Not all data need to be unlearned with equal priority
Understanding Unlearning Difficulty in LLMs
A neuro-inspired approach to selective knowledge removal
The Coreset Effect in LLM Unlearning
Why current unlearning benchmarks may be easier than they appear
Selective Forgetting in AI Models
A Novel Approach to Privacy-Compliant Unlearning
