Contrastive Covariance Subspaces for Activation Steering
cs.LG, cs.AI, cs.CL
Submitted: 2026-09-29
Updated: 2026-09-29
Terminology
Sources
- Faithful Bi-Directional Model Steering via Distribution Matching and Distributed Interchange Interventions
- Activation-Space Personality Steering: Hybrid Layer Selection for Stable Trait Control in LLMs
- Do Personality Traits Interfere? Geometric Limitations of Steering in Large Language Models
- Persona Vectors: Monitoring and Controlling Character Traits in Language Models
- Gemma 2: Improving Open Language Models at a Practical Size
- The Llama 3 Herd of Models
- Improving Activation Steering in Language Models with Mean-Centring
- Prompt-Activation Duality: Improving Activation Steering via Attention-Level Interventions
- Emotion Concepts and their Function in a Large Language Model
- Interpretable Steering of Large Language Models with Feature Guided Activation Additions
- Conceptors for Semantic Steering
- Steering Language Models With Activation Engineering
- Angular Steering: Behavior Control via Rotation in Activation Space
- AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders
- Qwen3 Technical Report
- Representation Engineering: A Top-Down Approach to AI Transparency
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: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks