# RuleSafe-VL: Evaluating Rule-Conditioned Decision Reasoning in Vision-Language Content Moderation

arXiv: https://arxiv.org/abs/2605.07760 · HTML: https://mlsanity.com/paper/2605.07760/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 33 references that are in the map.
- First posted on arXiv 8 May 2026.

## Builds on (most cited first, 13 of 13)

- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency](https://mlsanity.com/paper/2508.18265/) (2025)
- [The Hateful Memes Challenge: Detecting Hate Speech in Multimodal Memes](https://mlsanity.com/paper/2005.04790/) (2020)
- [MM-SafetyBench: A Benchmark for Safety Evaluation of Multimodal Large Language Models](https://mlsanity.com/paper/2311.17600/) (2023)
- [LlavaGuard: An Open VLM-based Framework for Safeguarding Vision Datasets and Models](https://mlsanity.com/paper/2406.05113/) (2024)
- [Evaluating GPT-3 Generated Explanations for Hateful Content Moderation](https://mlsanity.com/paper/2305.17680/) (2023)
- [GuardReasoner-VL: Safeguarding VLMs via Reinforced Reasoning](https://mlsanity.com/paper/2505.11049/) (2025)
- [MM-Soc: Benchmarking Multimodal Large Language Models in Social Media Platforms](https://mlsanity.com/paper/2402.14154/) (2024)
- [CASE-Bench: Context-Aware SafEty Benchmark for Large Language Models](https://mlsanity.com/paper/2501.14940/) (2025)
- [MSTS: A Multimodal Safety Test Suite for Vision-Language Models](https://mlsanity.com/paper/2501.10057/) (2025)
- [Enhanced Multimodal Content Moderation of Children's Videos using Audiovisual Fusion](https://mlsanity.com/paper/2405.06128/) (2024)
- [BLM-Guard: Explainable Multimodal Ad Moderation with Chain-of-Thought and Policy-Aligned Rewards](https://mlsanity.com/paper/2602.18193/) (2026)

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

- [Verdict Without the Rule: Diagnosing and Auditing Regulatory Rule Sensitivity in LLM Compliance Systems](https://mlsanity.com/paper/2610.12313/) (2026)
- [SemTree: Judge-Aware Semantic Decision Trees for Inspectable and Repairable Rule Decisions](https://mlsanity.com/paper/2610.13999/) (2026)

## Abstract (arXiv)

Platform content moderation applies explicit policy rules and context-dependent conditions to decide whether user content is allowed, restricted, or removed. A correct moderation outcome must therefore depend on which rules a case activates, how those rules interact, and whether the available evidence is sufficient. Current multimodal safety benchmarks largely reduce moderation to matching predefined final labels, leaving this underlying rule structure untested. As a result, a high benchmark score reveals little about whether a model applies the policy correctly or arrives at the correct label through superficial cues. To evaluate this rule-governed process, we introduce RuleSafe-VL, a benchmark for rule-conditioned decision reasoning in vision-language content moderation. Derived from publicly available platform moderation policies, RuleSafe-VL formalizes 93 atomic rules and 92 typed rule relations, yielding 2,166 context-sensitive image-text cases across three high-risk policy families. Its four diagnostic tasks decompose moderation into a rule-conditioned decision chain. They identify activated rules, recover rule interactions, judge decision sufficiency, and resolve outcomes once missing context is supplied. Experiments on 10 frontier, open-source, and safety-oriented VLMs reveal rule-relation recovery as the dominant bottleneck, where the best model reaches only 64.8 Macro-F1 and some safety-oriented models fall below 7 Macro-F1. Decision-state prediction also remains unreliable, peaking at 64.5 Macro-F1. RuleSafe-VL shifts moderation evaluation from final-label scoring toward diagnostic assessment of rule-conditioned decision reasoning.