Paul Pajo [Jan/3➞₿ ∎] #Insulin4All {#HODL} (@pageman)

Aman.ai의 'Top 30 papers' 목록과 Spotify 재생목록을 @NotebookLM으로 오디오 개요/설명 생성에 활용한 사례들을 언급하며, @cneuralnetwork의 papercode.in 같은 장난감(토이) 구현들이 실제로 유용하다는 관찰을 공유합니다. 연구 목록·오디오 해설·경량 구현을 통한 AI 정보 정리·실험적 도구 활용에 관한 논의입니다.

https://x.com/pageman/status/2013573560623829252

#notebooklm #aipapers #tools #papercode

Paul Pajo 🧢 [Jan/3➞₿ 🔑∎] #Insulin4All {#HODL} (@pageman) on X

@rohanpaul_ai @rheimann I've actually seen lists like https://t.co/r7jHdyEKAa ' s https://t.co/cTwAStD50T and even Spotify playlists that used @NotebookLM to create audio overviews / explainers but is only after I saw @cneuralnetwork 's https://t.co/P88HkBjOEP that I thought toy implementations of

X (formerly Twitter)

Tổng hợp các nghiên cứu RAG mới nhất ngày 25/12/23! Bao gồm các bài về MemR$^3$ cho tác nhân LLM, phát hiện ảo giác Faithfulness, Prompt Learning tăng cường truy xuất, suy luận đa bước, M$^3$KG-RAG và phân tích cảm xúc tài chính với RAG/LLM.

#RAG #LLM #AIResearch #NghiênCứuAI #AIPapers

https://www.reddit.com/r/LocalLLaMA/comments/1pugbsi/rag_paper_251223/

Yesterday I published a Jupyter Notebook which summarises individual PDF papers from arXiv (here: https://github.com/FahimF/fai-exp/blob/main/article_summarizer.ipynb)

Here’s a new implementation which summarises all new papers on arXiv for a given category. But 106 papers took about 2.5 hours to summarize 😛

You can see the results in the screenshot. And in case you want to do this yourself, the notebook and the code is here:

https://deepnote.com/@my-work-0e23/arXiv-9f1f011b-2520-4e9b-bda9-5c0919594ce3

#DeepLearning #Summarization #AIPapers

fai-exp/article_summarizer.ipynb at main · FahimF/fai-exp

Fast AI Practical Deep Learning for Coders experiments in Stable Diffusion - fai-exp/article_summarizer.ipynb at main · FahimF/fai-exp

GitHub