# SoFA: Shielded On-the-fly Alignment via Priority Rule Following

arXiv: https://arxiv.org/abs/2402.17358 · HTML: https://mlsanity.com/paper/2402.17358/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- Placed in LLM preference optimization methods, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 33 of its 53 references that are in the map.
- First posted on arXiv 27 Feb 2024.

## Placed in

- LLM alignment / RLHF / preference optimisation / [LLM preference optimization methods](https://mlsanity.com/alignment/llm-preference-optimization-methods/)

## Builds on (most cited first, 30 of 33)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [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)
- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [Self-Instruct: Aligning Language Models with Self-Generated Instructions](https://mlsanity.com/paper/2212.10560/) (2022)
- [Learning to summarize from human feedback](https://mlsanity.com/paper/2009.01325/) (2020)
- [The Rise and Potential of Large Language Model Based Agents: A Survey](https://mlsanity.com/paper/2309.07864/) (2023)
- [A General Language Assistant as a Laboratory for Alignment](https://mlsanity.com/paper/2112.00861/) (2021)
- [Red Teaming Language Models with Language Models](https://mlsanity.com/paper/2202.03286/) (2022)
- [Discovering Language Model Behaviors with Model-Written Evaluations](https://mlsanity.com/paper/2212.09251/) (2022)
- [Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2307.15217/) (2023)
- [Can Large Language Models Be an Alternative to Human Evaluations?](https://mlsanity.com/paper/2305.01937/) (2023)
- [BBQ: A Hand-Built Bias Benchmark for Question Answering](https://mlsanity.com/paper/2110.08193/) (2021)
- [Identifying the Risks of LM Agents with an LM-Emulated Sandbox](https://mlsanity.com/paper/2309.15817/) (2023)
- [Improving alignment of dialogue agents via targeted human judgements](https://mlsanity.com/paper/2209.14375/) (2022)
- [Reinforced Self-Training (ReST) for Language Modeling](https://mlsanity.com/paper/2308.08998/) (2023)
- [Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision](https://mlsanity.com/paper/2312.09390/) (2023)
- [Scalable agent alignment via reward modeling: a research direction](https://mlsanity.com/paper/1811.07871/) (2018)
- [Making Retrieval-Augmented Language Models Robust to Irrelevant Context](https://mlsanity.com/paper/2310.01558/) (2023)
- [Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Models](https://mlsanity.com/paper/2312.06585/) (2023)
- [Principle-Driven Self-Alignment of Language Models from Scratch with Minimal Human Supervision](https://mlsanity.com/paper/2305.03047/) (2023)
- [Risks from Learned Optimization in Advanced Machine Learning Systems](https://mlsanity.com/paper/1906.01820/) (2019)
- [Large Language Model Alignment: A Survey](https://mlsanity.com/paper/2309.15025/) (2023)
- [Exploring Large Language Models for Communication Games: An Empirical Study on Werewolf](https://mlsanity.com/paper/2309.04658/) (2023)
- [Scalable and Transferable Black-Box Jailbreaks for Language Models via Persona Modulation](https://mlsanity.com/paper/2311.03348/) (2023)
- [Fine-tuning language models to find agreement among humans with diverse preferences](https://mlsanity.com/paper/2211.15006/) (2022)
- [Tensor Trust: Interpretable Prompt Injection Attacks from an Online Game](https://mlsanity.com/paper/2311.01011/) (2023)

## Built on by (most cited first, 18 of 18)

- [A Survey on Self-Evolution of Large Language Models](https://mlsanity.com/paper/2404.14387/) (2024)
- [LLMs know their vulnerabilities: Uncover Safety Gaps through Natural Distribution Shifts](https://mlsanity.com/paper/2410.10700/) (2024)
- [Refuse Whenever You Feel Unsafe: Improving Safety in LLMs via Decoupled Refusal Training](https://mlsanity.com/paper/2407.09121/) (2024)
- [Caution for the Environment: Multimodal LLM Agents are Susceptible to Environmental Distractions](https://mlsanity.com/paper/2408.02544/) (2024)
- [IHEval: Evaluating Language Models on Following the Instruction Hierarchy](https://mlsanity.com/paper/2502.08745/) (2025)
- [A Closer Look at System Prompt Robustness](https://mlsanity.com/paper/2502.12197/) (2025)
- [SysBench: Can Large Language Models Follow System Messages?](https://mlsanity.com/paper/2408.10943/) (2024)
- [X-Boundary: Establishing Exact Safety Boundary to Shield LLMs from Multi-Turn Jailbreaks without Compromising Usability](https://mlsanity.com/paper/2502.09990/) (2025)
- [CantTalkAboutThis: Aligning Language Models to Stay on Topic in Dialogues](https://mlsanity.com/paper/2404.03820/) (2024)
- [LED-Merging: Mitigating Safety-Utility Conflicts in Model Merging with Location-Election-Disjoint](https://mlsanity.com/paper/2502.16770/) (2025)
- [LifeAlign: Lifelong Alignment for Large Language Models with Memory-Augmented Focalized Preference Optimization](https://mlsanity.com/paper/2509.17183/) (2025)
- [Baichuan Alignment Technical Report](https://mlsanity.com/paper/2410.14940/) (2024)
- [RuleAlign: Making Large Language Models Better Physicians with Diagnostic Rule Alignment](https://mlsanity.com/paper/2408.12579/) (2024)
- [Rule-Guided Feedback: Enhancing Reasoning by Enforcing Rule Adherence in Large Language Models](https://mlsanity.com/paper/2503.11336/) (2025)
- [Learning to Route: A Rule-Driven Agent Framework for Hybrid-Source Retrieval-Augmented Generation](https://mlsanity.com/paper/2510.02388/) (2025)
- [System Message Generation for User Preferences using Open-Source Models](https://mlsanity.com/paper/2502.11330/) (2025)
- [Structuring the Space of Sociotechnical Alignment](https://mlsanity.com/paper/2607.01250/) (2026)
- [Compliance, Capability, and Conflict: Benchmarking Multimodal LLMs under System Messages](https://mlsanity.com/paper/2608.19207/) (2026)

## Abstract (arXiv)

The alignment problem in Large Language Models (LLMs) involves adapting them to the broad spectrum of human values. This requirement challenges existing alignment methods due to diversity of preferences and regulatory standards. This paper introduces a novel alignment paradigm, priority rule following, which defines rules as the primary control mechanism in each dialog, prioritizing them over user instructions. Our preliminary analysis reveals that even the advanced LLMs, such as GPT-4, exhibit shortcomings in understanding and prioritizing the rules. Therefore, we present PriorityDistill, a semi-automated approach for distilling priority following signals from LLM simulations to ensure robust rule integration and adherence. Our experiments show that this method not only effectively minimizes misalignments utilizing only one general rule but also adapts smoothly to various unseen rules, ensuring they are shielded from hijacking and that the model responds appropriately.