# MDIT-Bench: Evaluating the Dual-Implicit Toxicity in Large Multimodal Models

arXiv: https://arxiv.org/abs/2505.17144 · HTML: https://mlsanity.com/paper/2505.17144/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in MLLM jailbreaks and adversarial robustness, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 20 of its 41 references that are in the map.
- First posted on arXiv 22 May 2025.

## Placed in

- Multimodal LLMs / vision-language models / [MLLM jailbreaks and adversarial robustness](https://mlsanity.com/mllm/mllm-jailbreaks-and-adversarial-robustness/)

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

- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://mlsanity.com/paper/2305.06500/) (2023)
- [Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone](https://mlsanity.com/paper/2404.14219/) (2024)
- [RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models](https://mlsanity.com/paper/2009.11462/) (2020)
- [SALAD-Bench: A Hierarchical and Comprehensive Safety Benchmark for Large Language Models](https://mlsanity.com/paper/2402.05044/) (2024)
- [Latent Hatred: A Benchmark for Understanding Implicit Hate Speech](https://mlsanity.com/paper/2109.05322/) (2021)
- [CogVLM2: Visual Language Models for Image and Video Understanding](https://mlsanity.com/paper/2408.16500/) (2024)
- [Exploring Hate Speech Detection in Multimodal Publications](https://mlsanity.com/paper/1910.03814/) (2019)
- [Toxicity Detection: Does Context Really Matter?](https://mlsanity.com/paper/2006.00998/) (2020)
- [CValues: Measuring the Values of Chinese Large Language Models from Safety to Responsibility](https://mlsanity.com/paper/2307.09705/) (2023)
- [Safety of Multimodal Large Language Models on Images and Texts](https://mlsanity.com/paper/2402.00357/) (2024)
- [Multimodal Situational Safety](https://mlsanity.com/paper/2410.06172/) (2024)
- [GOAT-Bench: Safety Insights to Large Multimodal Models through Meme-Based Social Abuse](https://mlsanity.com/paper/2401.01523/) (2024)
- [Parameter-Efficient and Student-Friendly Knowledge Distillation](https://mlsanity.com/paper/2205.15308/) (2022)
- [Queer In AI: A Case Study in Community-Led Participatory AI](https://mlsanity.com/paper/2303.16972/) (2023)
- [ToViLaG: Your Visual-Language Generative Model is Also An Evildoer](https://mlsanity.com/paper/2312.11523/) (2023)
- [Towards Robust Toxic Content Classification](https://mlsanity.com/paper/1912.06872/) (2019)
- [CommonIT: Commonality-Aware Instruction Tuning for Large Language Models via Data Partitions](https://mlsanity.com/paper/2410.03077/) (2024)

## Built on by (most cited first, 1 of 1)

- [SaFeR-VLM: Toward Safety-aware Fine-grained Reasoning in Multimodal Models](https://mlsanity.com/paper/2510.06871/) (2025)

## Abstract (arXiv)

The widespread use of Large Multimodal Models (LMMs) has raised concerns about model toxicity. However, current research mainly focuses on explicit toxicity, with less attention to some more implicit toxicity regarding prejudice and discrimination. To address this limitation, we introduce a subtler type of toxicity named dual-implicit toxicity and a novel toxicity benchmark termed MDIT-Bench: Multimodal Dual-Implicit Toxicity Benchmark. Specifically, we first create the MDIT-Dataset with dual-implicit toxicity using the proposed Multi-stage Human-in-loop In-context Generation method. Based on this dataset, we construct the MDIT-Bench, a benchmark for evaluating the sensitivity of models to dual-implicit toxicity, with 317,638 questions covering 12 categories, 23 subcategories, and 780 topics. MDIT-Bench includes three difficulty levels, and we propose a metric to measure the toxicity gap exhibited by the model across them. In the experiment, we conducted MDIT-Bench on 13 prominent LMMs, and the results show that these LMMs cannot handle dual-implicit toxicity effectively. The model's performance drops significantly in hard level, revealing that these LMMs still contain a significant amount of hidden but activatable toxicity. Data are available at https://github.com/nuo1nuo/MDIT-Bench.