# Steering Vector Fields for Context-Aware Inference-Time Control in Large Language Models

arXiv: https://arxiv.org/abs/2602.01654 · HTML: https://mlsanity.com/paper/2602.01654/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Placed in Activation steering methods, one of 13 directions in Mechanistic interpretability.
- Builds on 35 of its 48 references that are in the map.
- First posted on arXiv 2 Feb 2026.

## Placed in

- Mechanistic interpretability / [Activation steering methods](https://mlsanity.com/mechinterp/activation-steering-methods/)

## Builds on (most cited first, 30 of 35)

- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [FiLM: Visual Reasoning with a General Conditioning Layer](https://mlsanity.com/paper/1709.07871/) (2017)
- [Root Mean Square Layer Normalization](https://mlsanity.com/paper/1910.07467/) (2019)
- [Inference-Time Intervention: Eliciting Truthful Answers from a Language Model](https://mlsanity.com/paper/2306.03341/) (2023)
- [Steering Llama 2 via Contrastive Activation Addition](https://mlsanity.com/paper/2312.06681/) (2023)
- [Steering Language Models With Activation Engineering](https://mlsanity.com/paper/2308.10248/) (2023)
- [Discovering Language Model Behaviors with Model-Written Evaluations](https://mlsanity.com/paper/2212.09251/) (2022)
- [LEACE: Perfect linear concept erasure in closed form](https://mlsanity.com/paper/2306.03819/) (2023)
- [In-context Vectors: Making In Context Learning More Effective and Controllable Through Latent Space Steering](https://mlsanity.com/paper/2311.06668/) (2023)
- [AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders](https://mlsanity.com/paper/2501.17148/) (2025)
- [ReFT: Representation Finetuning for Language Models](https://mlsanity.com/paper/2404.03592/) (2024)
- [Analyzing the Generalization and Reliability of Steering Vectors](https://mlsanity.com/paper/2407.12404/) (2024)
- [Personalized Steering of Large Language Models: Versatile Steering Vectors Through Bi-directional Preference Optimization](https://mlsanity.com/paper/2406.00045/) (2024)
- [Aligning Large Language Models with Human Preferences through Representation Engineering](https://mlsanity.com/paper/2312.15997/) (2023)
- [LoFiT: Localized Fine-tuning on LLM Representations](https://mlsanity.com/paper/2406.01563/) (2024)
- [Understanding (Un)Reliability of Steering Vectors in Language Models](https://mlsanity.com/paper/2505.22637/) (2025)
- [CausalGym: Benchmarking causal interpretability methods on linguistic tasks](https://mlsanity.com/paper/2402.12560/) (2024)
- [Designing a Dashboard for Transparency and Control of Conversational AI](https://mlsanity.com/paper/2406.07882/) (2024)
- [A Unified Understanding and Evaluation of Steering Methods](https://mlsanity.com/paper/2502.02716/) (2025)
- [Multi-Attribute Steering of Language Models via Targeted Intervention](https://mlsanity.com/paper/2502.12446/) (2025)
- [On the Hardness of Faithful Chain-of-Thought Reasoning in Large Language Models](https://mlsanity.com/paper/2406.10625/) (2024)
- [Extending Activation Steering to Broad Skills and Multiple Behaviours](https://mlsanity.com/paper/2403.05767/) (2024)
- [Multi-property Steering of Large Language Models with Dynamic Activation Composition](https://mlsanity.com/paper/2406.17563/) (2024)
- [Advancing Parameter Efficiency in Fine-tuning via Representation Editing](https://mlsanity.com/paper/2402.15179/) (2024)
- [Towards Reliable Evaluation of Behavior Steering Interventions in LLMs](https://mlsanity.com/paper/2410.17245/) (2024)
- [M2IV: Towards Efficient and Fine-grained Multimodal In-Context Learning via Representation Engineering](https://mlsanity.com/paper/2504.04633/) (2025)
- [SAIF: A Sparse Autoencoder Framework for Interpreting and Steering Instruction Following of Language Models](https://mlsanity.com/paper/2502.11356/) (2025)
- [Taxonomy, Opportunities, and Challenges of Representation Engineering for Large Language Models](https://mlsanity.com/paper/2502.19649/) (2025)

## Built on by (most cited first, 13 of 13)

- [CreativityNeuro: Steering Language Model Weights to Improve Divergent Thinking and Reduce Mode Collapse](https://mlsanity.com/paper/2607.01433/) (2026)
- [ATLAS: Constitution-Conditioned Latent Geometry and Redistribution Across Language Models and Neural Perturbation Data](https://mlsanity.com/paper/2604.17663/) (2026)
- [When Is Rank-1 Steering Cheap? Geometry, Granularity, and Budgeted Search](https://mlsanity.com/paper/2605.16362/) (2026)
- [Does Steering Break Your Model? A Multi-Dimensional Evaluation Suite for LLM Steering Methods](https://mlsanity.com/paper/2610.07722/) (2026)
- [Temporal Preference Concepts and their Functions in a Large Language Model](https://mlsanity.com/paper/2606.05194/) (2026)
- [Closing the Loop: PID Feedback Control for Interpretable Activation Steering in Symbolic Music Generation](https://mlsanity.com/paper/2606.18790/) (2026)
- [Activation Steering Transfer to Agents: One Gain Ratio Does Not Identify Potency and Efficacy](https://mlsanity.com/paper/2607.09156/) (2026)
- [Inverted Detection and Control in Steering Vectors](https://mlsanity.com/paper/2608.02957/) (2026)
- [When Is a Task Vector Enough? An Empirical Theory of Implicit Multimodal ICL](https://mlsanity.com/paper/2608.13385/) (2026)
- [Semantic Fibers and Cross-Gram Interference: A Calculus of Safety Drift in Overcomplete Representations](https://mlsanity.com/paper/2609.14861/) (2026)
- [Test-Time Unlearning via Sparse Autoencoder](https://mlsanity.com/paper/2609.16229/) (2026)
- [Metacognitive Steering: Learning the Structure of Scientific Judgment](https://mlsanity.com/paper/2609.16245/) (2026)
- [Replay-Gated Neural Execution: Decoupling Persistent Behavioral Specifications from Neural Realizations in Frozen Language Models](https://mlsanity.com/paper/2609.22243/) (2026)

## Abstract (arXiv)

Steering vectors (SVs) offer a lightweight way to control large language models (LLMs) at inference time by shifting hidden activations, providing a practical middle ground between prompting and fine-tuning. Yet SVs can be unreliable in practice. Some concepts are unsteerable, and even when steering helps on average it can backfire for a non-trivial fraction of inputs. Reliability also degrades in long-form generation and multi-attribute steering. We take a geometric view of these failures. A static SV applies the same update vector everywhere in representation space, implicitly assuming that the concept-improving direction is constant across contexts. When the locally effective direction varies with the current activation, a single global vector can become misaligned, which yields weak or reversed effects. Guided by this perspective, we propose Steering Vector Fields (SVF), which learns a differentiable concept scoring function whose local gradient defines the steering direction at each activation, making interventions explicitly context-dependent. This formulation supports coordinated multi-layer interventions in a shared, aligned concept space, and enables efficient long-form and multi-attribute control within a unified framework. Across multiple LLMs and steering tasks, SVF delivers stronger and more reliable control, improving the practicality of inference-time steering.