Chain of Thought, PAL, ReAct

AI Discussions
1

Hello
I have gone through the research papers of COT, PAL and ReAct and these are the questions I have.

  1. So in COT do we fine tune a pretrained LLM with COT prompts ( I mean do we give a question followed by COT and then let the model generate the answer).

  2. At inference we are not providing any COT but the model is able to generate good answers for reasoning tasks.

  3. can you explain me how exactly we train/finetune using COT/PAL / ReAct.
    I mean what are inputs and outputs while we finetune/train the model like inputs are context+ COT prompts and outputs are answers (or) inputs are just context and outputs are COT+ answers.

2

You’re posting the same flavor of question in many different forum areas.
That’s typically not a successful method.

3

I am sorry for that. Will make sure not to repeat that next time.
Thank you