karin verspoor (professor)
5K posts
@karinv
Executive Dean @RMITComputing; AI in medicine #AAAiH; text mining, bioinformatics, #DigitalHealth #NLProc #techdiversity #TeamHB3 pronouns: she/her
You see this kind of thing is why I will never run out of content
Milindi Kodikara from @RMIT is presenting the long paper today at #ALTA2024 titled "Lesser the Shots, Higher the #Hallucinations: Exploration of Genetic Information Extraction using Generative Large Language Models", evaluating #LLMs on extraction of genetic information. #NLP
@JulietteBruce12 I have had one honest success using GPT (o1) to do research mathematics—I asked it for a counterexample to a strengthening of a conjecture I was thinking about, and it gave me a correct(!) counterexample.
People have been rewriting history and saying that "everyone has always believed that LLMs alone wouldn't be AGI and that extensive scaffolding around them would be necessary". No, throughout most of 2023 (the "sparks of AGI" era) the mainstream bay area belief was that LLMs were *already* AGI, and that merely scaling their parameter count and training data size by ~2 OOM without changing anything else would lead to super-intelligence.
Science is built on trust. What happens when someone destroys it? My first feature for @_TheTransmitter investigates the emotional and existential fallout for one lab in the aftermath of a misconduct case. thetransmitter.org/science-and-so…
You can Crawl entire website with Claude 3.5 or GPT4 with @firecrawl 💯 Its open-sourced and code in github - Turn entire websites into LLM-ready markdown or structured data. Scrape, crawl and extract with a single API. - Crawls all accessible subpages and give you clean data for each. No sitemap required. - The greatest benefit is that the extracted data is catered for LLM-based pipelines. - The api is self hostable and opensource ----- Some benefits of firecrawl 1. handles crawling (with or without sitemaps) 2. runs headless browsers scalably 3. handles bot protections and proxies 4. a team of dedicated engineers to solve the millions of edge-cases on the web for you 5. quality formatting to markdown by default Beautiful soup doesn't generalize, thats why we built firecrawl
