We are building pure AI generated newsletters to provide personalized, interactive and customized news reading experience for everyone.
And we are excited to release our first topic, the LLM newsletter (http://news.voker.io/), which aggregates news, opinions, new products, papers and all information related to LLM across multiple sources.
We are continue making better experience for readers, including but not limited to interactive reading, personalization and topic customization.
Hope this LLM newsletter can be helpful for everyone to learn what is happening everyday in this hot area! And follow our twitter (https://twitter.com/VokerNewsletter) to get the newest newsletters!
Any suggestions, feedbacks and requests for new topics are very appreciated!
Thank you for sharing. It’s really interesting, would you mind providing more technical details? Specifically, are you using open-source models? any Hugging Face tool? thanks
We are using a multi-agent architecture to source and generate the newsletter. Some key agents are:
- Sourcing agent: we fine-tune an OSS model to help understand the instruction about where to crawl the news and what information we’d like to add to the newsletter
- Ranking agent: pick up the most important and most relevant documents
- Generation agent: generation final text and image content of the newsletter.
There are multiple models involved in different agents, are for some agents like ranking and sourcing agents, we used huggingface pre-trained models.
Hope this can be helpful!
thanks @JingfeiDu that’s awesome.
Thank you for sharing.