Don't Forget Your Embeddings: Robust Knowledge Erasure via Precise Editing of Embeddings

arXiv:2606.03695 · cs.CL · Submitted 2026-06-02 · Read on arXiv

cs.CL

Submitted: 2026-06-02

Updated: 2026-08-30

Code: https://github.com/ClarSu/EMBER-Embedding-Erasure

Terminology

• Accuracy — 65× in this paper · explained in 1 episode
• Embeddings — 39× in this paper · explained in 2 episodes
• Robustness — 17× in this paper · explained in 5 episodes
• Matrix Factorization — 14× in this paper
• Sparsity — 14× in this paper · explained in 4 episodes
• Association for Computational Linguistics — 12× in this paper
• Transformer — 12× in this paper · explained in 2 episodes
• Adam — 7× in this paper · explained in 2 episodes
• Fine-Tuning — 6× in this paper · explained in 5 episodes
• Ablation — 5× in this paper · explained in 1 episode
• LLM judge — 4× in this paper · explained in 2 episodes
• Machine Unlearning — 4× in this paper · explained in 2 episodes
• Benchmark — 3× in this paper · explained in 2 episodes
• Feature Selection — 3× in this paper · explained in 4 episodes
• Interpretability — 3× in this paper · explained in 15 episodes
• Adversarial Prompting — 2× in this paper · explained in 1 episode
• Alignment — 2× in this paper · explained in 1 episode
• Reconstruction Error — 2× in this paper · explained in 2 episodes
• Trace — 2× in this paper · explained in 2 episodes
• Actor — 1× in this paper · explained in 3 episodes
• Adversarial Training — 1× in this paper · explained in 1 episode
• Aggregation — 1× in this paper · explained in 1 episode
• Approximate Unlearning — 1× in this paper · explained in 1 episode
• Decoupling — 1× in this paper · explained in 2 episodes
• Explainability — 1× in this paper · explained in 3 episodes
• Generalization — 1× in this paper · explained in 13 episodes
• Mechanistic interpretability — 1× in this paper · explained in 2 episodes
• Quantization — 1× in this paper · explained in 12 episodes
• Sparse Autoencoders (SAEs) — 1× in this paper · explained in 3 episodes
• Steering — 1× in this paper · explained in 2 episodes
• Tokenization — 1× in this paper · explained in 2 episodes

Sources

• Open Problems in Machine Unlearning for AI Safety
• Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
• Gemini: A Family of Highly Capable Multimodal Models
• Do Unlearning Methods Remove Information from Language Model Weights?
• Who's Harry Potter? Approximate Unlearning in LLMs
• The Llama 3 Herd of Models
• Word Meanings in Transformer Language Models
• Eight Methods to Evaluate Robust Unlearning in LLMs
• Parameter-Efficient Token Embedding Editing for Clinical Class-Level Unlearning
• Gemma 2: Improving Open Language Models at a Practical Size
• LLM Unlearning Should Be Form-Independent
• OpenAI GPT-5 System Card

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering