Training-Free Token-Level Steering for LLM Personalized Co-Writing

arXiv:2608.06069 · cs.CL · Submitted 2026-08-06 · Read on arXiv

Wenhao Mao, Chengbin Hou, Weixiao Wang, Jialiang Zhu, Min Liu, Yibin Hao, Hairong Lv

cs.CL

Submitted: 2026-08-06

Code: https://github.com/LengendaryHippopotamus/SteerWrite

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

Terminology

• RAG — 34× in this paper · explained in 3 episodes
• Steering — 23× in this paper · explained in 2 episodes
• Calibration — 14× in this paper · explained in 9 episodes
• Fine-Tuning — 8× in this paper · explained in 5 episodes
• Retrieval-Augmented Generation — 6× in this paper
• Evaluation Protocol — 5× in this paper · explained in 2 episodes
• Ablation — 4× in this paper · explained in 1 episode
• Benchmark — 4× in this paper · explained in 2 episodes
• Generalization — 4× in this paper · explained in 14 episodes
• Ablation Study — 3× in this paper · explained in 3 episodes
• Core — 3× in this paper · explained in 2 episodes
• Kernel Density Estimation — 3× in this paper
• Language Models (LLMs) — 3× in this paper · explained in 2 episodes
• Large Language Models (LLMs) — 3× in this paper · explained in 9 episodes
• Runtime Efficiency — 3× in this paper · explained in 2 episodes
• Training-Free Framework — 3× in this paper · explained in 2 episodes
• Cognitive Load — 2× in this paper · explained in 2 episodes
• Computational overhead — 2× in this paper · explained in 2 episodes
• Cosine Similarity — 2× in this paper · explained in 3 episodes
• Embeddings — 2× in this paper · explained in 2 episodes
• Foundation Model — 2× in this paper · explained in 3 episodes
• Foundation Models — 2× in this paper · explained in 8 episodes
• Papers — 2× in this paper · explained in 3 episodes
• Personalization — 2× in this paper · explained in 2 episodes
• Clarity — 1× in this paper · explained in 3 episodes
• Consistency — 1× in this paper · explained in 3 episodes
• Domain Adaptation — 1× in this paper · explained in 5 episodes
• Edge AI — 1× in this paper · explained in 2 episodes
• Evaluation Framework — 1× in this paper · explained in 3 episodes
• Generalization Capabilities — 1× in this paper · explained in 2 episodes
• Grounding — 1× in this paper · explained in 3 episodes
• Hallucination — 1× in this paper · explained in 4 episodes
• Human-in-the-loop — 1× in this paper · explained in 4 episodes
• In-Context Learning — 1× in this paper · explained in 7 episodes
• Inference-Time Guidance — 1× in this paper · explained in 3 episodes
• Latent Space — 1× in this paper · explained in 7 episodes
• Memorization — 1× in this paper · explained in 4 episodes
• Perplexity — 1× in this paper · explained in 3 episodes
• Regularization — 1× in this paper · explained in 3 episodes
• Retrieval-Augmented Generation (RAG) — 1× in this paper · explained in 21 episodes
• Robustness — 1× in this paper · explained in 6 episodes
• Semantic Alignment — 1× in this paper · explained in 6 episodes
• Sparsity — 1× in this paper · explained in 5 episodes
• Time To First Token — 1× in this paper
• Ultrasound Examination Reports — 1× in this paper

Sources

• Drift No More? Context Equilibria in Multi-Turn LLM Interactions
• Retrieval-Augmented Generation for Large Language Models: A Survey
• Fine-tuning and Utilization Methods of Domain-specific LLMs
• Generalization through Memorization: Nearest Neighbor Language Models
• Prefix-Tuning: Optimizing Continuous Prompts for Generation
• UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction
• Prototypical Human-AI Collaboration Behaviors from LLM-Assisted Writing in the Wild
• CodeNet: A Large-Scale AI for Code Dataset for Learning a Diversity of Coding Tasks
• Learning How to Ask: Querying LMs with Mixtures of Soft Prompts
• Selective Prompt Anchoring for Code Generation
• KNN-LM Does Not Improve Open-ended Text Generation
• Interactive Natural Language Processing
• Finetuned Language Models Are Zero-Shot Learners
• Medical Graph RAG: Towards Safe Medical Large Language Model via Graph Retrieval-Augmented Generation
• TAMA: A Human-AI Collaborative Thematic Analysis Framework Using Multi-Agent LLMs for Clinical Interviews
• Qwen3 Technical Report
• Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models
• Fine-tuning Large Language Models for Domain-specific Machine Translation

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