# PrismAgent: Illuminating Harm in Memes via a Zero-Shot Interpretable Multi-Agent Framework

arXiv: https://arxiv.org/abs/2605.02940 · HTML: https://mlsanity.com/paper/2605.02940/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Safety alignment, refusal and guardrails, one of 12 directions in LLM safety and security.
- Builds on 33 of its 54 references that are in the map.
- First posted on arXiv 1 May 2026.

## 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, 30 of 33)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [Universal and Transferable Adversarial Attacks on Aligned Language Models](https://mlsanity.com/paper/2307.15043/) (2023)
- [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://mlsanity.com/paper/2305.06500/) (2023)
- [VisualBERT: A Simple and Performant Baseline for Vision and Language](https://mlsanity.com/paper/1908.03557/) (2019)
- [ExpeL: LLM Agents Are Experiential Learners](https://mlsanity.com/paper/2308.10144/) (2023)
- [FLAVA: A Foundational Language And Vision Alignment Model](https://mlsanity.com/paper/2112.04482/) (2021)
- [The Hateful Memes Challenge: Detecting Hate Speech in Multimodal Memes](https://mlsanity.com/paper/2005.04790/) (2020)
- [AdaPlanner: Adaptive Planning from Feedback with Language Models](https://mlsanity.com/paper/2305.16653/) (2023)
- [MAGIS: LLM-Based Multi-Agent Framework for GitHub Issue Resolution](https://mlsanity.com/paper/2403.17927/) (2024)
- [LION : Empowering Multimodal Large Language Model with Dual-Level Visual Knowledge](https://mlsanity.com/paper/2311.11860/) (2023)
- [Detecting Harmful Memes and Their Targets](https://mlsanity.com/paper/2110.00413/) (2021)
- [Prompting for Multimodal Hateful Meme Classification](https://mlsanity.com/paper/2302.04156/) (2023)
- [Embodied Multi-Modal Agent trained by an LLM from a Parallel TextWorld](https://mlsanity.com/paper/2311.16714/) (2023)
- [Detecting and Understanding Harmful Memes: A Survey](https://mlsanity.com/paper/2205.04274/) (2022)
- [Hate-CLIPper: Multimodal Hateful Meme Classification based on Cross-modal Interaction of CLIP Features](https://mlsanity.com/paper/2210.05916/) (2022)
- [Detecting Hate Speech in Memes Using Multimodal Deep Learning Approaches: Prize-winning solution to Hateful Memes Challenge](https://mlsanity.com/paper/2012.12975/) (2020)
- [Multi-modal Agent Tuning: Building a VLM-Driven Agent for Efficient Tool Usage](https://mlsanity.com/paper/2412.15606/) (2024)
- [Pro-Cap: Leveraging a Frozen Vision-Language Model for Hateful Meme Detection](https://mlsanity.com/paper/2308.08088/) (2023)
- [Towards Low-Resource Harmful Meme Detection with LMM Agents](https://mlsanity.com/paper/2411.05383/) (2024)
- [Modularized Networks for Few-shot Hateful Meme Detection](https://mlsanity.com/paper/2402.11845/) (2024)
- [MIND: A Multi-agent Framework for Zero-shot Harmful Meme Detection](https://mlsanity.com/paper/2507.06908/) (2025)
- [Demystifying Hateful Content: Leveraging Large Multimodal Models for Hateful Meme Detection with Explainable Decisions](https://mlsanity.com/paper/2502.11073/) (2025)
- [Just KIDDIN: Knowledge Infusion and Distillation for Detection of INdecent Memes](https://mlsanity.com/paper/2411.12174/) (2024)

## Abstract (arXiv)

The rapid spread of memes makes harmful content detection increasingly crucial, as effective identification can curb the circulation of misinformation. However, existing methods rely heavily on high-volume annotated data, which leads to substantial training costs and limited generalization. To address these challenges, we propose PrismAgent, a zero-shot, multi-agent, interpretable framework. PrismAgent conceptualizes this task as a criminal case investigation, employing four specialized agents responsible for the analysis, investigation, prosecution, and judgment stages within a structured collaborative workflow. In the first stage, the analyst agent paraphrases each meme under benevolent and malicious assumptions to probe its underlying intent. The investigator agent then retrieves supporting evidence from an unannotated dataset and constructs contextual interpretations for the meme and its variants. Next, the prosecutor agent performs three independent preliminary judgments by pairing the original meme with each of the three interpretations. Finally, the judge agent deliberates across all evidence to render a final verdict. Moreover, PrismAgent's explicit multi-stage reasoning chain makes the model inherently interpretable, as every intermediate step is explicitly explained rather than only producing a final detection result. Extensive experiments on three public datasets show that PrismAgent significantly outperforms existing zero-shot detection methods.