How to Tame a Multi-Headed Hydra? Adaptive Multi-Category Safety Steering for Large Language Models

arXiv:2609.34514 · cs.CR, cs.CL, cs.LG · Submitted 2026-09-28 · Read on arXiv

cs.CR, cs.CL, cs.LG

Submitted: 2026-09-28

Updated: 2026-09-29

Code: https://github.com/mnmn-f/CAM-Steer

Terminology

• Steering — 161× in this paper · explained in 2 episodes
• Calibration — 17× in this paper · explained in 7 episodes
• Benchmark — 14× in this paper · explained in 2 episodes
• Activation Steering — 10× in this paper · explained in 4 episodes
• AUROC — 9× in this paper · explained in 2 episodes
• Ablation — 9× in this paper · explained in 1 episode
• International Conference on Learning Representations — 6× in this paper
• Alignment — 5× in this paper · explained in 1 episode
• Ablation study — 4× in this paper · explained in 1 episode
• Activation Addition — 4× in this paper · explained in 1 episode
• Contrastive Activation Addition — 4× in this paper
• National Alliance on Mental Illness — 4× in this paper
• Adaptive Safety — 3× in this paper · explained in 1 episode
• Robustness — 3× in this paper · explained in 5 episodes
• Transformer — 3× in this paper · explained in 2 episodes
• Core — 2× in this paper · explained in 2 episodes
• Information Leakage — 2× in this paper · explained in 2 episodes
• Large Language Models (LLMs) — 2× in this paper · explained in 6 episodes
• Accountability — 1× in this paper · explained in 2 episodes
• Adaptation — 1× in this paper · explained in 1 episode
• Adaptive Steering — 1× in this paper · explained in 1 episode
• Decode — 1× in this paper · explained in 2 episodes
• Elias — 1× in this paper · explained in 2 episodes
• Failure Modes — 1× in this paper · explained in 3 episodes
• Regularization — 1× in this paper · explained in 3 episodes
• Verifier — 1× in this paper · explained in 2 episodes

Sources

• Training Verifiers to Solve Math Word Problems
• ORBIT: Training-Free Multi-Attribute Behavioral Steering via Orthogonal Subspace Rotation
• Adaptive Multi-Subspace Representation Steering for Attribute Alignment in Large Language Models
• Gemma 2: Improving Open Language Models at a Practical Size
• The Llama 3 Herd of Models
• Qwen3 Technical Report
• Spherical Steering: Geometry-Aware Activation Rotation for Language Models

Related papers

• SoK: AI-Augmented Binary Reversing
• Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
• Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
• Efficient Fuzzy PSI under One-Sided Assumptions
• Sealing the Audit-Runtime Gap for LLM Skills
• Token Composition: A Graph Based on EVM Logs