Certain aspects and features of the present disclosure relate to providing contextually grounded recommendations using a large language model. For example, a method involves receiving domain specific data for a simulation and transforming the domain specific data into a labeled, natural language description of the domain specific data. The method also involves providing the labeled, natural language description and a classification task prompt with interaction history to a large language model (LLM) to generate a contextually enhanced LLM configured to produce context-aware output. The method further involves outputting, using the contextually enhanced LLM, an interactive list of scored actions corresponding to the simulation. The interactive list can be used to produce a sequence of actions to direct a process or control a machine.

Embodiments of this application disclose a method and an apparatus for adjusting a natural language sentence, and a storage medium. The method includes: receiving a natural language sentence including a placeholder, where the placeholder includes a type attribute, and the placeholder is occupied by a sentence constituent matching the type attribute; determining a position of the placeholder in the natural language sentence; and adding a preposition to the natural language sentence based on the type attribute of the placeholder and the position of the placeholder in the natural language sentence. The preposition is automatically added to the natural language sentence through the placeholder, so that readability and accuracy of the natural language sentence can be improved. In addition, when a word order of the natural language sentence is adjusted, the preposition is automatically updated correspondingly, which ensures the readability and the accuracy of the sentence.