Director, Learning Design & Development| PMIAsheville, NC, United States
With Generative AI, iteratively refining and optimizing prompts can lead to better AI-generated results. This may involve adjusting the specificity or clarity of the prompt to increase relevance and accuracy of results.
What examples do you have of how improving a prompt drastically changed the output quality? What specific changes did you make that led to the improvement?
There are framewoks to create prompt. This is part of the Prompt Desing discipline. Those that gave me and the initiatives where I was included are:R-T-F (Role-Task-Format), T-A-G (Task, action, goal), B-A-B (Before, after, bridge), C-A-R-E (context, action, result, example), R-I-S-E (role, input, steps, expectations).
It is quite helpful to use these methods. Thank you very much, I really appreciate it.
With Gen AI, refining prompts leads to better responses from the Gen AI's LLM. in an iterative refining way, the outcome becomes better after each iteration. Saving Changes...
Anonymous
Refining the prompt helps guide the AI model to give the desired results and as the conversation goes on, the AI model will continue to learn from the conversation and help refine the responses to exactly what the requester needs. Saving Changes...
The best way is to validate the output of Gen AI by ourselves using our trainings, skills and knowledge. Providing clear contexts and prompts will ensure that the output of AI will be aligned with our goals. Iterative refinement of prompts will help further strengthening the final outcome. Saving Changes...
I created a repetitive report, that at the beginning was of poor quality. Without haven't learned the techniques, I iterated to improve it. I uploaded a report template on how I wanted it. I also managed to interact with the LLM telling it to ask me seven questions one by one. Then I also asked it to write the report using formal, professional, direct yet amicable writing. I don't remember if I gave it a persona, but I probably did, as this is the first aspect I learned. I also told the LLM not to hallucinate. After several attempts, it worked better and gave me a report which I considered was at 80-90% of what I wanted. With the techniques I just learned, I believe the outputs will be better from now on. Saving Changes...
It's been a double edge sword. While refining prompts has produced a successful outcome, there are times when refining has been less than fruitful. On a few occasions, refining a prompt has created results less than ideal, forcing me to further refine. Saving Changes...
In my experience with GenAI, refining a prompt can completely transform the output. It's a great reminder that the quality of what you get depends so much on how you ask. Saving Changes...
Anonymous
aa Saving Changes...
Anonymous
Putting in the effort to assist AI at the beginning takes a while but the results are much more tailored to the task given. Being direct and specific has changed the output from a vague response into a great product that only required minor changes. Saving Changes...
Yong Wen JingSenior Consultant| Turner & TownsendSungai Petani, Kedah, Malaysia
It will definitely will enhance the outcome. I normally use it for drafting email. The content of my email will be for different audience. One of the biggest example is by just changing the target audience in my prompt, AI will generate 2 entire different content despite the the other criteria of the prompt is remain the same. Saving Changes...
PMI Team Member
