# Customize Multi-modal RAI Guardrails with Precedent-based predictions

arXiv: https://arxiv.org/abs/2507.20503 · HTML: https://mlsanity.com/paper/2507.20503/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Safety alignment, refusal and guardrails, one of 12 directions in LLM safety and security.
- Builds on 21 of its 29 references that are in the map.
- First posted on arXiv 28 Jul 2025.

## Placed in

- LLM safety and security / [Safety alignment, refusal and guardrails](https://mlsanity.com/safety/safety-alignment-refusal-and-guardrails/)

## Builds on (most cited first, 21 of 21)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [You Only Look Once: Unified, Real-Time Object Detection](https://mlsanity.com/paper/1506.02640/) (2015)
- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [LAION-400M: Open Dataset of CLIP-Filtered 400 Million Image-Text Pairs](https://mlsanity.com/paper/2111.02114/) (2021)
- [RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models](https://mlsanity.com/paper/2009.11462/) (2020)
- [Unsupervised Dense Information Retrieval with Contrastive Learning](https://mlsanity.com/paper/2112.09118/) (2021)
- [SGLang: Efficient Execution of Structured Language Model Programs](https://mlsanity.com/paper/2312.07104/) (2023)
- [Large Language Models Can Be Easily Distracted by Irrelevant Context](https://mlsanity.com/paper/2302.00093/) (2023)
- [MM-SafetyBench: A Benchmark for Safety Evaluation of Multimodal Large Language Models](https://mlsanity.com/paper/2311.17600/) (2023)
- [Red-Teaming the Stable Diffusion Safety Filter](https://mlsanity.com/paper/2210.04610/) (2022)
- [Not only Look, but also Listen: Learning Multimodal Violence Detection under Weak Supervision](https://mlsanity.com/paper/2007.04687/) (2020)
- [Unsafe Diffusion: On the Generation of Unsafe Images and Hateful Memes From Text-To-Image Models](https://mlsanity.com/paper/2305.13873/) (2023)
- [Challenges in Detoxifying Language Models](https://mlsanity.com/paper/2109.07445/) (2021)
- [Holistic Evaluation of Text-To-Image Models](https://mlsanity.com/paper/2311.04287/) (2023)
- [Can Machines Help Us Answering Question 16 in Datasheets, and In Turn Reflecting on Inappropriate Content?](https://mlsanity.com/paper/2202.06675/) (2022)
- [Safeguarding Large Language Models: A Survey](https://mlsanity.com/paper/2406.02622/) (2024)
- [Building Guardrails for Large Language Models](https://mlsanity.com/paper/2402.01822/) (2024)
- [UnsafeBench: Benchmarking Image Safety Classifiers on Real-World and AI-Generated Images](https://mlsanity.com/paper/2405.03486/) (2024)

## Built on by (most cited first, 2 of 2)

- [Beyond Text: Multimodal Jailbreaking of Vision-Language and Audio Models through Perceptually Simple Transformations](https://mlsanity.com/paper/2510.20223/) (2025)
- [PolicyShiftGuard: Benchmarking and Improving Policy-Adaptive Image Guardrails](https://mlsanity.com/paper/2607.05910/) (2026)

## Abstract (arXiv)

A multi-modal guardrail must effectively filter image content based on user-defined policies, identifying material that may be hateful, reinforce harmful stereotypes, contain explicit material, or spread misinformation. Deploying such guardrails in real-world applications, however, poses significant challenges. Users often require varied and highly customizable policies and typically cannot provide abundant examples for each custom policy. Consequently, an ideal guardrail should be scalable to the multiple policies and adaptable to evolving user standards with minimal retraining. Existing fine-tuning methods typically condition predictions on pre-defined policies, restricting their generalizability to new policies or necessitating extensive retraining to adapt. Conversely, training-free methods struggle with limited context lengths, making it difficult to incorporate all the policies comprehensively. To overcome these limitations, we propose to condition model's judgment on "precedents", which are the reasoning processes of prior data points similar to the given input. By leveraging precedents instead of fixed policies, our approach greatly enhances the flexibility and adaptability of the guardrail. In this paper, we introduce a critique-revise mechanism for collecting high-quality precedents and two strategies that utilize precedents for robust prediction. Experimental results demonstrate that our approach outperforms previous methods across both few-shot and full-dataset scenarios and exhibits superior generalization to novel policies.