Glad I subscribed earlier. I would be interested in the node part which seems to be the out of the box advantage versus Langchain. And the router with the retriever selection also is a great concept!
i believe it is this one: Lewis, Patrick, et al. "Retrieval-augmented generation for knowledge-intensive nlp tasks." Advances in Neural Information Processing Systems 33 (2020): 9459-9474. arxiv.org/abs/2005.11401
for what i have seen, smaller companies use notion a lot. but the bigger ones who have been using jira for years stick to confluence. but my experience is in the french market, so i´m not sure about elsewhere
i use both. you can't go wrong with either of them, I would say. and both communities are amazing. llamaindex has been doubling down on data parsing. you can use llama for document loading and langchain for the generation workflow. that's a great synergy
Glad I subscribed earlier. I would be interested in the node part which seems to be the out of the box advantage versus Langchain. And the router with the retriever selection also is a great concept!
AweSome!. Thanks for sharing by Jerry Liu and Alejandro.
he is such an interesting person. so great talking to him
As always, great content!
you're the best
maybe i should change the thumbnail 😓
@@alejandro_ao more clickbait
@@rifkyuntara6112 i'm trying to make it more clickable 😅, not super good at it!
great talk!!
So relatable, particularly initial over-reach and problem discovery. Solution depth is incredible. What an amazing human.
he really is
20:00 Is anyone able to find the link to the original RAG paper?
i believe it is this one:
Lewis, Patrick, et al. "Retrieval-augmented generation for knowledge-intensive nlp tasks." Advances in Neural Information Processing Systems 33 (2020): 9459-9474.
arxiv.org/abs/2005.11401
@@alejandro_ao nice, my boss is the author of that paper
@@zuowang5185 let's have him on the podcast!
Congrats. Awesome vídeo
i appreciate it!
Great stuff, always a pleasure hearing Jerry Liu speak from LlamaIndex! Hope you check out our new multi-agents and AI teams release! :)
You can use specialized models for unstructured data extraction for a fraction of the cost
llamaparse is free for up to 1000 pages / day
like col pali? or what did you have in mind?
Fantastic content
Much appreciated :)
Is Notion beating Atlassian Confluence?
for what i have seen, smaller companies use notion a lot. but the bigger ones who have been using jira for years stick to confluence. but my experience is in the french market, so i´m not sure about elsewhere
LlamaIndex vs LangChain?
i use both. you can't go wrong with either of them, I would say. and both communities are amazing. llamaindex has been doubling down on data parsing. you can use llama for document loading and langchain for the generation workflow. that's a great synergy
LlamaParse is okay with tables but struggles with intricate mathematical formulas.
Guess its hard to do formulas for everyone.