# EVADE-Bench: Multimodal Benchmark for Evaluating and Enhancing Evasive Content Detection

arXiv: https://arxiv.org/abs/2505.17654 · HTML: https://mlsanity.com/paper/2505.17654/

## Key facts

- Cited by 5 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 21 of its 37 references that are in the map.
- First posted on arXiv 23 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, 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)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Universal and Transferable Adversarial Attacks on Aligned Language Models](https://mlsanity.com/paper/2307.15043/) (2023)
- [A Survey of Large Language Models](https://mlsanity.com/paper/2303.18223/) (2023)
- [A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions](https://mlsanity.com/paper/2311.05232/) (2023)
- [The Hateful Memes Challenge: Detecting Hate Speech in Multimodal Memes](https://mlsanity.com/paper/2005.04790/) (2020)
- [LongVILA: Scaling Long-Context Visual Language Models for Long Videos](https://mlsanity.com/paper/2408.10188/) (2024)
- [Multimodal Foundation Models: From Specialists to General-Purpose Assistants](https://mlsanity.com/paper/2309.10020/) (2023)
- [InFoBench: Evaluating Instruction Following Ability in Large Language Models](https://mlsanity.com/paper/2401.03601/) (2024)
- [A Survey of Attacks on Large Vision-Language Models: Resources, Advances, and Future Trends](https://mlsanity.com/paper/2407.07403/) (2024)
- [Beyond the Limits: A Survey of Techniques to Extend the Context Length in Large Language Models](https://mlsanity.com/paper/2402.02244/) (2024)
- [LogicBench: Towards Systematic Evaluation of Logical Reasoning Ability of Large Language Models](https://mlsanity.com/paper/2404.15522/) (2024)
- [A Comprehensive Survey of Large Language Models and Multimodal Large Language Models in Medicine](https://mlsanity.com/paper/2405.08603/) (2024)
- [A Survey on Large Language Models with some Insights on their Capabilities and Limitations](https://mlsanity.com/paper/2501.04040/) (2025)
- [Prism: A Framework for Decoupling and Assessing the Capabilities of VLMs](https://mlsanity.com/paper/2406.14544/) (2024)
- [Can't See the Forest for the Trees: Benchmarking Multimodal Safety Awareness for Multimodal LLMs](https://mlsanity.com/paper/2502.11184/) (2025)
- [Beyond Utility: Evaluating LLM as Recommender](https://mlsanity.com/paper/2411.00331/) (2024)
- [Investigating LLM Applications in E-Commerce](https://mlsanity.com/paper/2408.12779/) (2024)
- [SWE2: SubWord Enriched and Significant Word Emphasized Framework for Hate Speech Detection](https://mlsanity.com/paper/2409.16673/) (2024)
- [Deep Learning Approaches for Detecting Adversarial Cyberbullying and Hate Speech in Social Networks](https://mlsanity.com/paper/2406.17793/) (2024)

## Built on by (most cited first, 5 of 5)

- [RuCL: Stratified Rubric-Based Curriculum Learning for Multimodal Large Language Model Reasoning](https://mlsanity.com/paper/2602.21628/) (2026)
- [Structuring Reasoning for Complex Rules Beyond Flat Representations](https://mlsanity.com/paper/2510.05134/) (2025)
- [Yuvion VL: A Multimodal Foundation Model for Adversarial Content and AI Safety](https://mlsanity.com/paper/2606.25034/) (2026)
- [Visual Compliance via Executable Safety Rule Entailment](https://mlsanity.com/paper/2609.18328/) (2026)
- [SemTree: Judge-Aware Semantic Decision Trees for Inspectable and Repairable Rule Decisions](https://mlsanity.com/paper/2610.13999/) (2026)

## Abstract (arXiv)

E-commerce platforms increasingly rely on Large Language Models (LLMs) and Vision Language Models (VLMs) to detect illicit or misleading product content. However, these models remain vulnerable to evasive content, which refers to inputs that have been deliberately modified through techniques such as word splitting, euphemistic language, or image cropping to conceal policy violations while still conveying prohibited claims. Crucially, detecting such content requires a model to simultaneously master two capabilities: accurately comprehending complex rules, and correctly inferring the true intent behind deliberately obfuscated multimodal inputs. While prior work has separately explored LLM reasoning over complex rules and LLM-based detection of evasive content, no existing benchmark combines both within a unified evaluation framework. This gap is particularly consequential in e-commerce, where accurate moderation demands that both capabilities operate in concert. To address this gap, we introduce EVADE-Bench, the first expert-curated Chinese multimodal benchmark specifically designed to evaluate LLMs and VLMs on evasive content detection in real-world e-commerce scenarios. Our comprehensive evaluation of 26 open- and closed-source LLMs and VLMs reveals that even state-of-the-art models frequently misclassify evasive samples. We further demonstrate that clearer rule categorization significantly improves model prediction consistency and reduces false predictions, highlighting the critical role of benchmark design in enabling reliable evaluation. To explore paths for performance improvement, we investigate the feasibility of multi-agent decomposition for multimodal reasoning, wherein visual description and logical inference are decoupled into separate agents, and find that this strategy yields notable accuracy gains.