# Cautious Context Steering for Language Model Personalization

arXiv: https://arxiv.org/abs/2608.05813 · HTML: https://mlsanity.com/paper/2608.05813/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 20 of its 27 references that are in the map.
- First posted on arXiv 6 Aug 2026.

## Builds on (most cited first, 20 of 20)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2307.15217/) (2023)
- [Personalized Soups: Personalized Large Language Model Alignment via Post-hoc Parameter Merging](https://mlsanity.com/paper/2310.11564/) (2023)
- [A Roadmap to Pluralistic Alignment](https://mlsanity.com/paper/2402.05070/) (2024)
- [Personalizing Reinforcement Learning from Human Feedback with Variational Preference Learning](https://mlsanity.com/paper/2408.10075/) (2024)
- [The PRISM Alignment Dataset: What Participatory, Representative and Individualised Human Feedback Reveals About the Subjective and Multicultural Alignment of Large Language Models](https://mlsanity.com/paper/2404.16019/) (2024)
- [On Releasing Annotator-Level Labels and Information in Datasets](https://mlsanity.com/paper/2110.05699/) (2021)
- [Democratizing Large Language Models via Personalized Parameter-Efficient Fine-tuning](https://mlsanity.com/paper/2402.04401/) (2024)
- [PersonalLLM: Tailoring LLMs to Individual Preferences](https://mlsanity.com/paper/2409.20296/) (2024)
- [LoRe: Personalizing LLMs via Low-Rank Reward Modeling](https://mlsanity.com/paper/2504.14439/) (2025)
- [Language Model Personalization via Reward Factorization](https://mlsanity.com/paper/2503.06358/) (2025)
- [Moral Machine or Tyranny of the Majority?](https://mlsanity.com/paper/2305.17319/) (2023)
- [Context Steering: Controllable Personalization at Inference Time](https://mlsanity.com/paper/2405.01768/) (2024)
- [FaST: Feature-aware Sampling and Tuning for Personalized Preference Alignment with Limited Data](https://mlsanity.com/paper/2508.04698/) (2025)
- [Swap-guided Preference Learning for Personalized Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2603.12595/) (2026)
- [The Reward Model Selection Crisis in Personalized Alignment](https://mlsanity.com/paper/2512.23067/) (2025)

## Abstract (arXiv)

Personalizing language models (LMs) to individual user preferences is essential for aligning responses with diverse goals and backgrounds. Existing methods typically train a separate adapter for each user or learn a reward model whose scores depend on the user. Despite explicitly optimizing for each user, these methods must learn from limited observations and therefore suffer from data sparsity and poor generalization to unseen users and domains. In-context learning (ICL) and Context Steering (CoS) can instead provide more effective personalization by conditioning the base LM directly on user context and leveraging its pretrained capabilities without per-user training. Yet neither adapts the influence of that context across decoding steps: ICL leaves it uncontrolled, whereas CoS applies a fixed steering coefficient and requires two LM forward passes per step. We propose Cautious Context Steering (CCS), which adds a lightweight adapter to a frozen backbone LM to decide at each token whether and how strongly user context should affect generation. The adapter learns this behavior from an oracle context-conditioned LM and preserves the base LM when the context is not helpful. A single CCS adapter trained on only one dataset improves generation quality both in-domain and across four out-of-distribution personalization benchmarks, demonstrating robust generalization to new users and domains. CCS also avoids per-user fine-tuning and the additional context-conditioned forward pass required by CoS, substantially reducing inference cost.