LACUNA: A Testbed for Evaluating Localization Precision for LLM Unlearning
cs.CL, cs.AI, cs.LG
Submitted: 2026-07-02
Updated: 2026-09-25
Terminology
Sources
- Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
- Gradient Routing: Masking Gradients to Localize Computation in Neural Networks
- Who's Harry Potter? Approximate Unlearning in LLMs
- Eternal Sunshine of the Spotless Net: Selective Forgetting in Deep Networks
- BLUR: A Benchmark for LLM Unlearning Robust to Forget-Retain Overlap
- Training Data Leakage Analysis in Language Models
- Analyzing Leakage of Personally Identifiable Information in Language Models
- 2 OLMo 2 Furious
- Olmo 3
- How Data Inter-connectivity Shapes LLMs Unlearning: A Structural Unlearning Perspective
- REBEL: Hidden Knowledge Recovery via Evolutionary-Based Evaluation Loop
- PANORAMA: A synthetic PII-laced dataset for studying sensitive data memorization in LLMs
- Beyond Data Filtering: Knowledge Localization for Capability Removal in LLMs
- Unlearning Isn't Deletion: Investigating Reversibility of Machine Unlearning in LLMs
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