# Simulating Hate Speech Cascades with Multi-LLM Agents: Empirical Grounding, Modeling Fidelity, and Intervention Strategies

arXiv: https://arxiv.org/abs/2606.18264 · HTML: https://mlsanity.com/paper/2606.18264/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Social simulation and multi-agent debate, one of 16 directions in LLM agents and tool use.
- Builds on 10 of its 29 references that are in the map.
- First posted on arXiv 21 May 2026.

## Placed in

- LLM agents and tool use / [Social simulation and multi-agent debate](https://mlsanity.com/agents/social-simulation-and-multi-agent-debate/)

## Builds on (most cited first, 10 of 10)

- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [Out of One, Many: Using Language Models to Simulate Human Samples](https://mlsanity.com/paper/2209.06899/) (2022)
- [Automated Hate Speech Detection and the Problem of Offensive Language](https://mlsanity.com/paper/1703.04009/) (2017)
- [Large Language Models Empowered Agent-based Modeling and Simulation: A Survey and Perspectives](https://mlsanity.com/paper/2312.11970/) (2023)
- [Directions in Abusive Language Training Data: Garbage In, Garbage Out](https://mlsanity.com/paper/2004.01670/) (2020)
- [Spread of hate speech in online social media](https://mlsanity.com/paper/1812.01693/) (2018)
- [Epidemic processes in complex networks](https://mlsanity.com/paper/1408.2701/) (2014)
- [Characterizing and Detecting Hateful Users on Twitter](https://mlsanity.com/paper/1803.08977/) (2018)
- [The Effect of Extremist Violence on Hateful Speech Online](https://mlsanity.com/paper/1804.05704/) (2018)
- [Social Influence and Unfollowing Accelerate the Emergence of Echo Chambers](https://mlsanity.com/paper/1905.03919/) (2019)

## Abstract (arXiv)

Faithful modeling of hateful content propagation on online platforms remains an open problem for moderation research. Classical cascade models that do not explicitly represent the profile, community, and content factors associated with hateful-content propagation may yield moderation strategies that behave less effectively when deployed in real-world scenarios. Multi-agent large language model (LLM) systems can, in principle, make each reshare decision depend on the user's profile, the surrounding community, and the post's content, but it remains unclear whether this added flexibility actually reproduces real hateful cascades more faithfully than classical baselines. We study three hateful Bluesky cascades and a size-matched benign control. In the empirical Bluesky data, we found that: 97.4--99.7% of reposters take a hostile stance; toxicity-engagement homophily is higher on the diffusion tree than on the follower graph for hateful cascades; topology is star-like for the hateful cascades (most reposts come directly from the root) versus tree-like for the benign cascade (reposts propagate through multi-hop chains). In simulation, a multi-LLM-agent simulator reproduces the stance monoculture and the toxicity-delta direction. A structured ablation identifies agent heterogeneity as the leading fidelity factor, and amplifier targeting on dense networks yields 7.5--12.9% reduction at 5.7% benign collateral.