# Now You See the Hate: Adaptive View Retrieval for Hidden Hateful Illusions

arXiv: https://arxiv.org/abs/2607.19061 · HTML: https://mlsanity.com/paper/2607.19061/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 26 references that are in the map.
- First posted on arXiv 21 Jul 2026.

## Builds on (most cited first, 17 of 17)

- [Intriguing Properties of Vision Transformers](https://mlsanity.com/paper/2105.10497/) (2021)
- [Better Aggregation in Test-Time Augmentation](https://mlsanity.com/paper/2011.11156/) (2020)
- [Visual Anagrams: Generating Multi-View Optical Illusions with Diffusion Models](https://mlsanity.com/paper/2311.17919/) (2023)
- [Grounding Visual Illusions in Language: Do Vision-Language Models Perceive Illusions Like Humans?](https://mlsanity.com/paper/2311.00047/) (2023)
- [A Survey of Multimodal Hallucination Evaluation and Detection](https://mlsanity.com/paper/2507.19024/) (2025)
- [Diffusion Illusions: Hiding Images in Plain Sight](https://mlsanity.com/paper/2312.03817/) (2023)
- [The Illusion-Illusion: Vision Language Models See Illusions Where There Are None](https://mlsanity.com/paper/2412.18613/) (2024)
- [IllusionCAPTCHA: A CAPTCHA based on Visual Illusion](https://mlsanity.com/paper/2502.05461/) (2025)
- [Illusory VQA: Benchmarking and Enhancing Multimodal Models on Visual Illusions](https://mlsanity.com/paper/2412.08169/) (2024)
- [Do VLMs Perceive or Recall? Probing Visual Perception vs. Memory with Classic Visual Illusions](https://mlsanity.com/paper/2601.22150/) (2026)
- [Evaluating Vision-Language Models on Bistable Images](https://mlsanity.com/paper/2405.19423/) (2024)
- [SemVink: Advancing VLMs' Semantic Understanding of Optical Illusions via Visual Global Thinking](https://mlsanity.com/paper/2506.02803/) (2025)
- [Hate in Plain Sight: On the Risks of Moderating AI-Generated Hateful Illusions](https://mlsanity.com/paper/2507.22617/) (2025)
- [Do Large Vision-Language Models Distinguish between the Actual and Apparent Features of Illusions?](https://mlsanity.com/paper/2506.05765/) (2025)
- [Illusions in Humans and AI: How Visual Perception Aligns and Diverges](https://mlsanity.com/paper/2508.12422/) (2025)
- [Do vision models perceive illusory motion in static images like humans?](https://mlsanity.com/paper/2604.09853/) (2026)
- [A Framework for Generating Semantically Ambiguous Images to Probe Human and Machine Perception](https://mlsanity.com/paper/2603.24730/) (2026)

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

- [ControlTrace: Recovering Control Fields for Hidden-Content Recognition](https://mlsanity.com/paper/2609.34807/) (2026)

## Abstract (arXiv)

Hateful optical illusions expose a serious gap in current multimodal safety systems. On original-view hateful illusions, previous work shows that six moderation classifiers achieve at most 20.9 to 24.5% accuracy and nine state-of-the-art VLMs remain at or below 10.2% with illusion-aware prompting, leaving most hidden hate undetected. We formulate hidden hateful illusion detection as a perceptual retrieval problem and propose Adaptive View Retrieval. This retrieve-and-calibrate framework assembles a complementary view bank for the image and hidden-message templates, adaptively selects which views to trust, retrieves hidden-message identities, and calibrates whether the recovered evidence is harmful. On HatefulIllusion with a frozen CLIP encoder, Adaptive View Retrieval reaches 93.2% balanced accuracy on the held-out test split. It substantially outperforms original-view baselines and fixed single-transform filters across hate slangs, hate symbols, and visibility levels. The same design also surpasses official fine-tuned CLIP baselines, matches or exceeds human performance on IllusionMNIST, IllusionFashionMNIST, and IllusionAnimals, and outperforms zoom-out preprocessing on HC-Bench under the SemVink protocol. Together, these results show that robust multimodal moderation requires recovering hidden meaning before deciding whether it is harmful.