arXiv AI
CANS: Capability-Preserving Neuron Steering for Large Language Models
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
Hugging Face Trending Papers
2d ago
How Far Can You Steer? Behavior-Capability Frontiers in Contrastive Activation Addition
Activation steering can modify language-model behavior without parameter updates, but behavioral efficacy alone does not determine whether an intervention is usable. Contrastive Activation Addition (C...
arXiv Machine Learning
1d ago
How Far Can You Steer? Behavior-Capability Frontiers in Contrastive Activation Addition
arXiv:2611.06001v1 Announce Type: new Abstract: Activation steering can modify language-model behavior without parameter updates, but behavioral efficacy alone does not determine whether an intervent...
arXiv Computation and Language
1d ago
Steer2Edit: From Activation Steering to Component-Level Editing
arXiv:2602.09870v3 Announce Type: replace Abstract: Steering methods influence Large Language Model behavior by identifying semantic directions in hidden representations, but are typically realized t...
arXiv Computation and Language
Sep 3
GAPS: Dimension-Level Gates for Conditional Activation Steering
The paper introduces GAPS, a dimension‑level gating approach for activation steering in language models. GAPS uses two training‑free gates—a static separability gate based on AUROC and a dynamic posterior gate based on a Gaussian model—to selectively apply steering vectors only to neurons that carry reliable concept information or are currently mis‑activated. Experiments on Gemma‑3 and Qwen‑3 show that GAPS improves or matches the performance of token‑level methods, notably reducing Gemma‑3’s toxicity rate from 6.52% to 0.48% under a fixed capability budget.
Hugging Face Trending Papers
Sep 1
GAPS: Dimension-Level Gates for Conditional Activation Steering
Activation steering suppresses undesired behaviors in language models by adding a steering vector to the hidden state during generation. Recent conditional methods such as CAST and DSAS improve the be...
arXiv AI
Jul 29
Minimizing Targeted Activations: Input-Only Suppression of Evaluation-Awareness Latents in Large Language Models
arXiv:2607. 25907v1 Announce Type: cross Abstract: Activation steering controls model behavior by editing internal activations at inference time.