Jonathan Birch (@jbcloudcentre) profile picture

A visionary leader, strategist, entrepreneur and technologist.

London UK
Joined August 2016
Photo by Jonathan Birch (@jbcloudcentre)Photo by Jonathan Birch (@jbcloudcentre)Photo by Jonathan Birch (@jbcloudcentre)Photo by Jonathan Birch (@jbcloudcentre)Photo by Jonathan Birch (@jbcloudcentre)Photo by Jonathan Birch (@jbcloudcentre)Photo by Jonathan Birch (@jbcloudcentre)Photo by Jonathan Birch (@jbcloudcentre)Photo by Jonathan Birch (@jbcloudcentre)Photo by Jonathan Birch (@jbcloudcentre)
Jonathan Birch retweeted
A Novel RAG Approach That Understands The Whole Document Context RAG has rapidly evolved to be the standard way to apply LLMs in production. However, most methods are still limited because most existing methods retrieve only short contiguous chunks from a retrieval corpus, limiting holistic understanding of the overall document context. A new approach named RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval (link in alt) proposes recursively embedding, clustering, and summarizing chunks of text, constructing a tree with differing levels of summarization from the bottom up. At inference time, the RAPTOR model retrieves from this tree, integrating information across lengthy documents at different levels of abstraction. Controlled experiments show that retrieval with recursive summaries offers significant improvements over traditional retrieval-augmented LMs on several tasks. On question-answering tasks that involve complex, multi-step reasoning, we show state-of-the-art results; for example, by coupling RAPTOR retrieval with the use of GPT-4, we can improve the best performance on the QuALITY benchmark by 20% in absolute accuracy. We will continue to see more methods like this one that are designed to improve document understanding and RAG. The good thing about this method is that it focuses on the retriever piece, while most of the other methods tinker with the context by adding noise or web results.
30
262
26
1,288
213,108
Jonathan Birch retweeted
Walmart used an LLM (PaLM-2) generated embeddings for their DLRM to make product search better on walmart.com They are about to push this out into their live website because the results were so good. Noteworthy other firms have already done this such as Google
12
87
5
560
109,873
Last day to apply for this innovative program #inSurTech #AWS #Iot Learn more & apply here: go.aws/3Gba8AL
1
25
Nivda GPUs power chat GPT, their blogs have many great insights into #generativeai #insuretech #5g #iot https://teapawt.net/t.co/D3qhuzBNMz
1
137
Encouraging to see good satisfaction scores from those that use digital FNOl as reported in the JD Power 2022 US claims-digital experience-study. buff.ly/3YvvLU1 #insurance #insurtech
1
36
Has to be easy use so Building MLGUI, user interfaces for machine learning applications is critical #insurance #insurtec #telco venturebeat.com/2021/07/19/b… via @VentureBeat
We must change the way we talk about women being attacked or abused by men. Words Matter.
We can all change the way we talk about it.
Global insurance M&A activity set to soar in 2021 intelligentinsurer.com/news/…
I registered for Underwriting Innovation USA. Check it out here underwriting-innovation-usa.…
Wife of Broadway actor Nick Cordero says he has died following long battle with COVID-19. He was 41 years old. instagram.com/p/CCSBM89Axt_/…
357
1,395
364
8,898
IOT temperature and moisture sensors monitor high value wine collections...Great idea. #insurtech #insurance https://teapawt.net/t.co/6UgkZRkiRD
Interesting 90 day plan to accelerate Digital particularly for #insurers #insurance who are called for lowest adaption of Digital. mck.co/2WZ0dH5 via @McKinsey
APIs unlock Digital #insurance #Insurtech https://teapawt.net/t.co/0X43wT0nfv