Attention Manifolds: Steering or Blocking Language Models by Editing Learned B-Spline Surfaces
cs.LG
Submitted: 2026-09-24
Updated: 2026-09-24
Terminology
Sources
- GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints
- KAN: Kolmogorov-Arnold Networks
- Kolmogorov-Arnold Attention: Is Learnable Attention Better For Vision Transformers?
- The Llama 3 Herd of Models
- InKAN: B-Spline KANs via Truncated Power Form
- Nested Attention: Semantic-aware Attention Values for Concept Personalization
- Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free
- 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