# System Report for CCL25-Eval Task 10: Prompt-Driven Large Language Model Merge for Fine-Grained Chinese Hate Speech Detection

arXiv: https://arxiv.org/abs/2512.09563 · HTML: https://mlsanity.com/paper/2512.09563/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 14 of its 20 references that are in the map.
- First posted on arXiv 10 Dec 2025.

## Builds on (most cited first, 14 of 14)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Don't Stop Pretraining: Adapt Language Models to Domains and Tasks](https://mlsanity.com/paper/2004.10964/) (2020)
- [TIES-Merging: Resolving Interference When Merging Models](https://mlsanity.com/paper/2306.01708/) (2023)
- [Merging Models with Fisher-Weighted Averaging](https://mlsanity.com/paper/2111.09832/) (2021)
- [HateXplain: A Benchmark Dataset for Explainable Hate Speech Detection](https://mlsanity.com/paper/2012.10289/) (2020)
- [Learning from the Worst: Dynamically Generated Datasets to Improve Online Hate Detection](https://mlsanity.com/paper/2012.15761/) (2020)
- [Latent Hatred: A Benchmark for Understanding Implicit Hate Speech](https://mlsanity.com/paper/2109.05322/) (2021)
- [Model Breadcrumbs: Scaling Multi-Task Model Merging with Sparse Masks](https://mlsanity.com/paper/2312.06795/) (2023)
- [Toxicity Detection: Does Context Really Matter?](https://mlsanity.com/paper/2006.00998/) (2020)
- [COLD: A Benchmark for Chinese Offensive Language Detection](https://mlsanity.com/paper/2201.06025/) (2022)
- [Detecting Harmful Content On Online Platforms: What Platforms Need Vs. Where Research Efforts Go](https://mlsanity.com/paper/2103.00153/) (2021)
- [Leveraging Intra-User and Inter-User Representation Learning for Automated Hate Speech Detection](https://mlsanity.com/paper/1804.03124/) (2018)
- [STATE ToxiCN: A Benchmark for Span-level Target-Aware Toxicity Extraction in Chinese Hate Speech Detection](https://mlsanity.com/paper/2501.15451/) (2025)

## Abstract (arXiv)

The proliferation of hate speech on Chinese social media poses urgent societal risks, yet traditional systems struggle to decode context-dependent rhetorical strategies and evolving slang. To bridge this gap, we propose a novel three-stage LLM-based framework: Prompt Engineering, Supervised Fine-tuning, and LLM Merging. First, context-aware prompts are designed to guide LLMs in extracting implicit hate patterns. Next, task-specific features are integrated during supervised fine-tuning to enhance domain adaptation. Finally, merging fine-tuned LLMs improves robustness against out-of-distribution cases. Evaluations on the STATE-ToxiCN benchmark validate the framework's effectiveness, demonstrating superior performance over baseline methods in detecting fine-grained hate speech.