Hidden not Deleted: How Networks Suppress Entangled Features
cs.LG, cs.AI
Submitted: 2026-09-23
Updated: 2026-09-23
Terminology
Sources
- Hidden Progress in Deep Learning: SGD Learns Parities Near the Computational Limit
- Sparse Autoencoders Find Highly Interpretable Features in Language Models
- Gaussian Error Linear Units (GELUs)
- The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning
- TOFU: A Task of Fictitious Unlearning for LLMs
- Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets
- Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small
- Unlearning Isn't Deletion: Investigating Reversibility of Machine Unlearning in LLMs
- Erase or Hide? Suppressing Spurious Unlearning Neurons for Robust Unlearning
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks