In this task, we are editing machine-translated contexts, queries and responses that are user interactions with LLMs. An interaction pair in this context would normally consist of:
Context: Background/reference information that the user’s prompt is based on.
Prompt or Query: The user’s first turn in a conversation. It can be a question, request, or instruction that guides the AI’s response.
Response: The bot’s response to the user.
User: “customer” and a “customer service representative”, the person that uses the model to perform or automate a given task.
Model or bot: The bot side of a human-to-chatbot conversation, i.e. the LLM response.
1.2 Task details
Do not under any circumstances use a generative chat/LLM model such as ChatGPT at any stage of the process. Do not use the model to create content, check facts, brainstorm ideas, etc.
In this task, you will receive prompts that have been machine translated.
You will need to post edit these texts according to the localization guidelines below. Keep an eye out for locale-specific instances that should be updated (e.g. dates) and language equivalents
