Large language models $($LLMs$)$ are trained on massive datasets of text and code. What are the drawbacks of LLMs that applications using LangChain framework aims to solve?

To ensure that the data is free of errors and duplicates.

To answer the queries that are beyond the date of the training data.

To enable LLMs to be aware of organizational internal data so that it can be used to build organization specific applications.

To improve the efficiency of the LLM$\text{'}$s training process.