The rise of large language models has brought about significant advancements in the field of natural language processing. However, these models often have the potential to generate content that can be hallucinatory or toxic. To this end, we organize NLPCC 2024 Shared Task 10, i.e., Regulating Large Language Models, which includes two sub-tasks: Multimodal Hallucination Detection for Multimodal Large Language Models and Detoxifying Large Language Models. In the first task, we construct a fine-grained and human-calibrated benchmark for multimodal hallucination detection, named MHaluBench, which contains 1270 training data, 600 validation data and 300 test data. The second task draws on the SafeEdit benchmark, containing 4050 training data, 2700 validation data and 540 test data. The aim is to design and implement strategies to prevent large language models from generating toxic content. This paper presents details of the shared tasks, datasets, evaluation metrics and evaluation results.