Deployable Per-Instance Multi-Layer Activation Steering for Large Language Models

arXiv:2608.08829 · cs.CL, cs.AI · Submitted 2026-08-09 · Read on arXiv

Muhammad Faishal Adly Nelwan, Alfan Farizki Wicaksono

cs.CL, cs.AI

Submitted: 2026-08-09

Updated: 2026-08-11

Comments: 43 pages, 24 figures, 30 tables. Under review at ACL Rolling Review (August 2026)

Code: https://github.com/pesolosep/per-instance-layer-steering

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Steering — 64× in this paper · explained in 2 episodes
• Perplexity — 21× in this paper · explained in 3 episodes
• AUC — 10× in this paper · explained in 2 episodes
• Activation Steering — 9× in this paper · explained in 5 episodes
• Accuracy — 4× in this paper · explained in 1 episode
• Accountability — 3× in this paper · explained in 2 episodes
• Activation Addition — 3× in this paper · explained in 1 episode
• Coherence — 3× in this paper · explained in 2 episodes
• Papers — 3× in this paper · explained in 3 episodes
• Additivity — 2× in this paper · explained in 1 episode
• Calibration — 2× in this paper · explained in 7 episodes
• Contrastive Activation Addition — 2× in this paper · explained in 2 episodes
• Core — 2× in this paper · explained in 2 episodes
• Embeddings — 2× in this paper · explained in 2 episodes
• Trace — 2× in this paper · explained in 2 episodes
• Abstention — 1× in this paper · explained in 1 episode
• Adam — 1× in this paper · explained in 2 episodes
• Cosine Similarity — 1× in this paper · explained in 3 episodes
• Cross-Model Transfer — 1× in this paper · explained in 2 episodes
• Diffusion Model — 1× in this paper · explained in 10 episodes
• Diffusion Models — 1× in this paper · explained in 9 episodes
• Editor — 1× in this paper · explained in 2 episodes
• Elias — 1× in this paper · explained in 2 episodes
• Empirical Methods in Natural Language Processing — 1× in this paper
• Entanglement — 1× in this paper · explained in 4 episodes
• Foundation Model — 1× in this paper · explained in 3 episodes
• Foundation Models — 1× in this paper · explained in 7 episodes
• Hallucination — 1× in this paper · explained in 4 episodes
• International Conference on Learning Representations — 1× in this paper
• Interpretability — 1× in this paper · explained in 15 episodes
• Mechanistic Interpretability — 1× in this paper · explained in 4 episodes
• Pruning — 1× in this paper · explained in 4 episodes
• Representation Engineering — 1× in this paper · explained in 2 episodes
• Selection Bias — 1× in this paper · explained in 4 episodes
• Transformer — 1× in this paper · explained in 2 episodes

Sources

• Aya Expanse: Combining Research Breakthroughs for a New Multilingual Frontier
• The Effectiveness of Style Vectors for Steering Large Language Models: A Human Evaluation
• Where to Steer: Input-Dependent Layer Selection for Steering Improves LLM Alignment
• The Llama 3 Herd of Models
• To Steer or Not to Steer? Mechanistic Error Reduction with Abstention for Language Models
• Programming Refusal with Conditional Activation Steering
• Inference-Time Intervention: Eliciting Truthful Answers from a Language Model
• Activation-Space Personality Steering: Hybrid Layer Selection for Stable Trait Control in LLMs
• Multi-Attribute Steering of Language Models via Targeted Intervention
• Beyond Linear Steering: Unified Multi-Attribute Control for Language Models
• Analyzing the Generalization and Reliability of Steering Vectors
• Learning to Steer: Input-dependent Steering for Multimodal LLMs
• Steering Language Models With Activation Engineering
• Discovering Language Model Behaviors with Model-Written Evaluations
• Angular Steering: Behavior Control via Rotation in Activation Space
• Adaptive Activation Steering: A Tuning-Free LLM Truthfulness Improvement Method for Diverse Hallucinations Categories
• Steering Llama 2 via Contrastive Activation Addition
• Automating Steering for Safe Multimodal Large Language Models
• Controlling Language and Diffusion Models by Transporting Activations
• Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering