# In-Distribution Steering: Balancing Control and Coherence in Language Model Generation

arXiv: https://arxiv.org/abs/2510.13285 · HTML: https://mlsanity.com/paper/2510.13285/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Placed in Activation steering methods, one of 13 directions in Mechanistic interpretability.
- Builds on 19 of its 37 references that are in the map.
- First posted on arXiv 15 Oct 2025.

## Placed in

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

## Builds on (most cited first, 19 of 19)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Understanding intermediate layers using linear classifier probes](https://mlsanity.com/paper/1610.01644/) (2016)
- [Representation Engineering: A Top-Down Approach to AI Transparency](https://mlsanity.com/paper/2310.01405/) (2023)
- [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)
- [Refusal in Language Models Is Mediated by a Single Direction](https://mlsanity.com/paper/2406.11717/) (2024)
- [Steering Language Models With Activation Engineering](https://mlsanity.com/paper/2308.10248/) (2023)
- [AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedback](https://mlsanity.com/paper/2305.14387/) (2023)
- [The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasets](https://mlsanity.com/paper/2310.06824/) (2023)
- [Large Language Model Alignment: A Survey](https://mlsanity.com/paper/2309.15025/) (2023)
- [Reinforcement Learning for LLM Post-Training: A Survey](https://mlsanity.com/paper/2407.16216/) (2024)
- [Interpreting the Curse of Dimensionality from Distance Concentration and Manifold Effect](https://mlsanity.com/paper/2401.00422/) (2023)
- [The Blessing and Curse of Dimensionality in Safety Alignment](https://mlsanity.com/paper/2507.20333/) (2025)
- [Self-AMPLIFY: Improving Small Language Models with Self Post Hoc Explanations](https://mlsanity.com/paper/2402.12038/) (2024)
- [Towards Achieving Concept Completeness for Textual Concept Bottleneck Models](https://mlsanity.com/paper/2502.11100/) (2025)

## Built on by (most cited first, 9 of 9)

- [Local Linearity of LLMs Enables Activation Steering via Model-Based Linear Optimal Control](https://mlsanity.com/paper/2604.19018/) (2026)
- [Steering at the Source: Style Modulation Heads for Robust Persona Control](https://mlsanity.com/paper/2603.13249/) (2026)
- [Steer Like the LLM: Activation Steering that Mimics Prompting](https://mlsanity.com/paper/2605.03907/) (2026)
- [Activation Steering for Aligned Open-ended Generation without Sacrificing Coherence](https://mlsanity.com/paper/2604.08169/) (2026)
- [Prefill-Time Intervention for Mitigating Hallucination in Large Vision-Language Models](https://mlsanity.com/paper/2604.25642/) (2026)
- [Tethered Reasoning: Decoupling Entropy from Hallucination in Quantized LLMs via Manifold Steering](https://mlsanity.com/paper/2602.17691/) (2026)
- [Closing the Loop: PID Feedback Control for Interpretable Activation Steering in Symbolic Music Generation](https://mlsanity.com/paper/2606.18790/) (2026)
- [Steering Interference Reflects the Model's Defaults, Not the Behavior Directions](https://mlsanity.com/paper/2609.06951/) (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)

Activation steering methods control large language model (LLM) behavior by modifying internal activations at inference time. However, most existing activation steering methods rely on a fixed steering strength, leading to either insufficient control or unadapted intervention that degrades text plausibility and coherence. We introduce In-Distribution Steering (IDS), a novel method that adapts steering strength based on the input data distribution in representation space. IDS dynamically adjusts interventions according to how far a given input lies within the distribution, enabling adaptive intervention and generation stability during text generation. Experiments demonstrate that IDS achieves strong accuracy on classification tasks while producing coherent text without collapse, making IDS particularly well suited for real-world applications.