mlsanityData to 14 Oct 2026

Attribute-Based Activation Steering of LLMs for Group-Specific Explanation Generation

arXiv 2608.29215 ↗First posted 29 Aug 2026cs.CLcited by 0

On the map: Mechanistic interpretability → Activation steering methods

Abstract (arXiv)

To effectively enable people to understand new topics, explanations should be tailored to their backgrounds and abilities. Prompting alone has been shown to be insufficient for creating such explanations, and other computational methods are missing so far. Therefore, this paper investigates whether LLMs can be steered to generate explanations that are tailored to a specific group of people. To this end, we propose an approach that first identifies group-specific attributes in terms of explanatory style and knowledge of a specific target group. Building on activation engineering, it then computes attribute-based steering vectors and adds them to the internal activations of an LLM during inference to enable a fine-grained steering. In our experiments, we assess the steering effectiveness in terms of specificity and factuality of the generated explanations. Additionally, we evaluate the explanations in a study with human experts from different target groups. Compared to prompting and state-of-the-art steering baselines, our approach tailors the explanations significantly better to the target group while maintaining the best specificity-factuality balance.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Activation steering methods, one of 13 directions in Mechanistic interpretability. Builds on 33 of its 62 references that are in the map. First posted on arXiv 29 Aug 2026.

Builds on

33 of 62 references.

References · 33

  1. 2023Sparse Autoencoders Find Highly Interpretable Features in Language Modelscited by 1,420
  2. 2023FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generationcited by 1,282
  3. 2025Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Modelscited by 1,261
  4. 2017Explanation in Artificial Intelligence: Insights from the Social Sciencescited by 1,206
  5. 2019Plug and Play Language Models: A Simple Approach to Controlled Text Generationcited by 799
  6. 2019ELI5: Long Form Question Answeringcited by 575
  7. 2020KILT: a Benchmark for Knowledge Intensive Language Taskscited by 573
  8. 2018Explaining Explanations: An Overview of Interpretability of Machine Learningcited by 508
  9. 2018e-SNLI: Natural Language Inference with Natural Language Explanationscited by 468
  10. 2019Explain Yourself! Leveraging Language Models for Commonsense Reasoningcited by 367
  11. 2025Persona Vectors: Monitoring and Controlling Character Traits in Language Modelscited by 345
  12. 2020Deep Learning for Text Style Transfer: A Surveycited by 162
21 more
  1. 2022Towards Faithful Model Explanation in NLP: A Surveycited by 130
  2. 2022Towards Human-centered Explainable AI: A Survey of User Studies for Model Explanationscited by 85
  3. 2023Sensitivity, Performance, Robustness: Deconstructing the Effect of Sociodemographic Promptingcited by 81
  4. 2024Adaptive Activation Steering: A Tuning-Free LLM Truthfulness Improvement Method for Diverse Hallucinations Categoriescited by 77
  5. 2020One Explanation Does Not Fit All: The Promise of Interactive Explanations for Machine Learning Transparencycited by 64
  6. 2024Exploring LLM Prompting Strategies for Joint Essay Scoring and Feedback Generationcited by 63
  7. 2024Style Vectors for Steering Generative Large Language Modelcited by 56
  8. 2025Multi-Attribute Steering of Language Models via Targeted Interventioncited by 46
  9. 2020SemEval-2020 Task 4: Commonsense Validation and Explanationcited by 42
  10. 2023Explainable Claim Verification via Knowledge-Grounded Reasoning with Large Language Modelscited by 37
  11. 2023Learning Interpretable Style Embeddings via Prompting LLMscited by 30
  12. 2024User Embedding Model for Personalized Language Promptingcited by 28
  13. 2025Beyond Linear Steering: Unified Multi-Attribute Control for Language Modelscited by 25
  14. 2024Style-Specific Neurons for Steering LLMs in Text Style Transfercited by 24
  15. 2023Know Your Audience: Do LLMs Adapt to Different Age and Education Levels?cited by 16
  16. 2021Belief-based Generation of Argumentative Claimscited by 14
  17. 2025Embedding-to-Prefix: Parameter-Efficient Personalization for Pre-Trained Large Language Modelscited by 10
  18. 2019Towards Content Transfer through Grounded Text Generationcited by 9
  19. 2023P5: Plug-and-Play Persona Prompting for Personalized Response Selectioncited by 8
  20. 2024Attribute Controlled Fine-tuning for Large Language Models: A Case Study on Detoxificationcited by 7
  21. 2022"Mama Always Had a Way of Explaining Things So I Could Understand'': A Dialogue Corpus for Learning to Construct Explanationscited by 6

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025ExpertSteer: Intervening in LLMs through Expert Knowledge
  2. 2026Causally Grounded Mechanistic Interpretability for LLMs with Faithful Natural-Language Explanations
  3. 2025SteerX: Disentangled Steering for LLM Personalization
  4. 2023XplainLLM: A Knowledge-Augmented Dataset for Reliable Grounded Explanations in LLMs
  5. 2023Logic-Scaffolding: Personalized Aspect-Instructed Recommendation Explanation Generation using LLMs
  6. 2026From Plausible to Actionable: A Position on LLM Self-Explanations
  7. 2026Rational Synthesizers or Heuristic Followers? Analyzing LLMs in RAG-based Question-Answering
  8. 2025Interpreting and Steering LLMs with Mutual Information-based Explanations on Sparse Autoencoders