Awesome! Looks like AgentWorker / Runner architecture does a similar job of query decomposition, searches with tools for the replies to sub queries, and combines it all in a final report. How do these classes and cases relate?
Glad to see this demo! SQQE is favorite query engine. Does this workflow pass the context used to answer each sub-question to the LLM for the final synthesis? I thought SQQE was doing that. (Not sure about the comment in notebook that SQQE can’t use multiple tools. I’m using it with a set of tools and it chooses where to send each sub-question.) Definitely interested in trying out this technique, due to ability to customize the workflow.
Sir I'm working on my final year project there are 2 main modules in it 1) previous year paper detailed analysis system along with sample paper generation as per trends ,and study roadmap provider 2) notes generation module from textbook content I'm confused what to use where .. whether fine tuned llm , or RAG or anything other ? Can you please explain, it is for engineering students (1st -4th semester, each one has 6 subjects ), there are 7 different branches.
Sir I'm working on my final year project there are 2 main modules in it 1) previous year paper detailed analysis system along with sample paper generation as per trends ,and study roadmap provider 2) notes generation module from textbook content I'm confused what to use where .. whether fine tuned llm , or RAG or anything other ? Can you please explain, it is for engineering students (1st -4th semester, each one has 6 subjects ), there are 7 different branches.
How to break up questions that have multiple parts, where understanding/answering one part affects how you form questions for the succeeding parts and answer these parts.
A demo on Sub question query engine with Multi-Modal RAG and/or queries could be super useful.
This is super dope! I can’t wait to create my own example! You guys are constantly dropping knowledge and amazing demo’s! Thank you so much!
wow, very cool!
Awesome. Could you please share notebook in the tutorial?
yes, please!!
Whoops! We've just added it to the description.
@@LlamaIndex thanks!!
Is there a way to get the answers to the question by conversing with the user?
Could you please share notebook in the tutorial? Not able to see any notebook link on description thanks..
Awesome!
Looks like AgentWorker / Runner architecture does a similar job of query decomposition, searches with tools for the replies to sub queries, and combines it all in a final report.
How do these classes and cases relate?
Glad to see this demo! SQQE is favorite query engine. Does this workflow pass the context used to answer each sub-question to the LLM for the final synthesis? I thought SQQE was doing that. (Not sure about the comment in notebook that SQQE can’t use multiple tools. I’m using it with a set of tools and it chooses where to send each sub-question.) Definitely interested in trying out this technique, due to ability to customize the workflow.
Sir I'm working on my final year project there are 2 main modules in it
1) previous year paper detailed analysis system along with sample paper generation as per trends ,and study roadmap provider
2) notes generation module from textbook content
I'm confused what to use where .. whether fine tuned llm , or RAG or anything other ?
Can you please explain, it is for engineering students (1st -4th semester, each one has 6 subjects ), there are 7 different branches.
Sir I'm working on my final year project there are 2 main modules in it
1) previous year paper detailed analysis system along with sample paper generation as per trends ,and study roadmap provider
2) notes generation module from textbook content
I'm confused what to use where .. whether fine tuned llm , or RAG or anything other ?
Can you please explain, it is for engineering students (1st -4th semester, each one has 6 subjects ), there are 7 different branches.
How to break up questions that have multiple parts, where understanding/answering one part affects how you form questions for the succeeding parts and answer these parts.