mlsanityData to 14 Oct 2026

Predicting Steering Vectors and Adapter Weights for Few-Shot Author-Style Transfer

arXiv 2610.03163 ↗First posted 2 Oct 2026cs.CL cs.AIcited by 0

On the map: Mechanistic interpretability → Activation steering methods

Abstract (arXiv)

Adapting large language models to an individual author's style from a few examples is challenging, and scientific writing sharpens the difficulty: formal conventions leave little surface variation, and authors write about their own topics, so extracted ``style'' easily entangles with content. We study style-conditioned abstract generation from a few example abstracts per author and propose three methods: (1) contrastive activation steering, (2) a network that predicts steering vectors, and (3) a hypernetwork that predicts LoRA adapters. We find a consistent trade-off between style imitation and output quality: fine-tuning buys most of the available style signal but forfeits fluency, while the hypernetwork achieves the best trade-off on both seen and unseen authors. Our steering operates at author level, contrasting an author's abstracts against style-neutral generations for the same content. This holds topic fixed, removes the need for a predefined style inventory, and outperforms inventory-based steering. % [EDIT 1a] softened "no single optimal axis" claim Moreover, our analyses demonstrate that manually extracted and predicted steering vectors are near-orthogonal yet score comparably, indicating that style conditioning here can admit at least two unrelated directions rather than requiring one particular axis.

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 20 of its 38 references that are in the map. First posted on arXiv 2 Oct 2026.

Builds on

20 of 38 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025Personalized Text Generation with Contrastive Activation Steering7 Mar
  2. 2023Steering Llama 2 via Contrastive Activation Addition9 Dec

Other references · 18

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2019The Curious Case of Neural Text Degenerationcited by 2,916
  3. 2025gpt-oss-120b & gpt-oss-20b Model Cardcited by 1,386
  4. 2023LaMP: When Large Language Models Meet Personalizationcited by 437
  5. 2023Large Language Models are Effective Text Rankers with Pairwise Ranking Promptingcited by 307
  6. 2024Do Llamas Work in English? On the Latent Language of Multilingual Transformerscited by 281
  7. 2022Same Author or Just Same Topic? Towards Content-Independent Style Representationscited by 86
  8. 2020Look at the First Sentence: Position Bias in Question Answeringcited by 84
  9. 2024Personalized Pieces: Efficient Personalized Large Language Models through Collaborative Effortscited by 74
  10. 2025Text-to-LoRA: Instant Transformer Adaptioncited by 58
  11. 2024StyleDistance: Stronger Content-Independent Style Embeddings with Synthetic Parallel Examplescited by 35
  12. 2025Middle-Layer Representation Alignment for Cross-Lingual Transfer in Fine-Tuned LLMscited by 29
6 more
  1. 2025Controlling Large Language Models Through Concept Activation Vectorscited by 28
  2. 2021Control Prefixes for Parameter-Efficient Text Generationcited by 24
  3. 2025Instant Personalized Large Language Model Adaptation via Hypernetworkcited by 18
  4. 2024TinyStyler: Efficient Few-Shot Text Style Transfer with Authorship Embeddingscited by 17
  5. 2023Understanding writing style in social media with a supervised contrastively pre-trained transformercited by 12
  6. 2024P-React: Synthesizing Topic-Adaptive Reactions of Personality Traits via Mixture of Specialized LoRA Expertscited by 7

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026AuthorMix: Modular Authorship Style Transfer via Layer-wise Adapter Mixing
  2. 2026Can Activation Steering Capture Multidimensional Authorship Style?
  3. 2025Steering Large Language Models for Machine Translation Personalization
  4. 2025Capturing Classic Authorial Style in Long-Form Story Generation with GRPO Fine-Tuning
  5. 2026On the Limits of Steering Vectors for Preference-Aligned Generation
  6. 2025Personalized Image Generation from an Author Writing Style
  7. 2023ParaGuide: Guided Diffusion Paraphrasers for Plug-and-Play Textual Style Transfer
  8. 2026From Profiles to Steering Vectors: Global Sparse Priors and Local Semantic Calibration for Personalized Text Generation