What Does It Mean to Forget a Person? Individual-Level Unlearning in Vision-Language Models

arXiv:2609.33481 · cs.CR, cs.CY · Submitted 2026-09-27 · Read on arXiv

cs.CR, cs.CY

Submitted: 2026-09-27

Updated: 2026-09-27

Terminology

• Benchmark — 43× in this paper · explained in 2 episodes
• Identity Access — 22× in this paper
• Identity Binding — 16× in this paper
• Fine-Tuning — 14× in this paper · explained in 5 episodes
• Attribute Access — 13× in this paper
• Identity Reconstruction — 10× in this paper
• Machine Unlearning — 7× in this paper · explained in 2 episodes
• Gradient Ascent — 5× in this paper
• Alignment — 4× in this paper · explained in 1 episode
• Pruning — 4× in this paper · explained in 4 episodes
• Accuracy — 3× in this paper · explained in 1 episode
• Adam — 3× in this paper · explained in 2 episodes
• Benchmarking — 3× in this paper · explained in 3 episodes
• Negative Preference Optimization — 3× in this paper
• Adaptation — 2× in this paper · explained in 1 episode
• Trace — 2× in this paper · explained in 2 episodes
• Consistency — 1× in this paper · explained in 3 episodes
• Data Provenance — 1× in this paper · explained in 2 episodes
• Instruction Tuning — 1× in this paper · explained in 2 episodes
• Modality Gap — 1× in this paper · explained in 4 episodes
• Provenance — 1× in this paper · explained in 4 episodes
• Reasoning Traces — 1× in this paper · explained in 3 episodes
• Transformer — 1× in this paper · explained in 2 episodes
• Verification — 1× in this paper · explained in 2 episodes
• Vision-Language Models (VLMs) — 1× in this paper · explained in 6 episodes

Sources

• Qwen3-VL Technical Report
• The Llama 3 Herd of Models
• GPT-4o System Card
• TOFU: A Task of Fictitious Unlearning for LLMs
• MUSE: Machine Unlearning Six-Way Evaluation for Language Models
• PEBench: A Fictitious Dataset to Benchmark Machine Unlearning for Multimodal Large Language Models
• Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning

Related papers

• SoK: AI-Augmented Binary Reversing
• Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
• Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
• Efficient Fuzzy PSI under One-Sided Assumptions
• Sealing the Audit-Runtime Gap for LLM Skills
• Token Composition: A Graph Based on EVM Logs