Hacker Newsnew | past | comments | ask | show | jobs | submit | MrCoffee7's commentslogin

GLM-4.5 from https://chat.z.ai because it renders math notation so well that other LLMs tend to mangle and it uses color coding when showing Python code.

There is a simpler explanation of the cited article here "Los Alamos team finds a new path toward quantum machine learning" https://www.lanl.gov/media/news/0724-quantum-machine-learnin...

Some of the stuff at the bleeding edge though like gene therapy is so expensive at millions of dollars per patient that health insurance companies are not covering it, so that stuff is not actually worthwhile to that many patients. The vaccine tech is covered by health insurance.

The A-Z sort option does not seem to work at all?

This awesome-math Github repo has a large list of many learning resources https://github.com/rossant/awesome-math



Thanks! Since https://www.wsj.com/us-news/law/american-sentenced-to-8-year... appears hardwalled and no one has supplied an archive link, I've switched the top URL to this.


Your tool looks like a great start, but I think it could use more types of search filters. A vibe type filter for the kind of tone the blog has would be useful, eg. educational, entertainment, provoking, relaxing, etc. A screenshot of part of their home screen shown on the info card for the blog would also help show the visual asthetic and vibe of the blog. It would also be nice to have an exclusion filter if you don't want to see certain subjects or people. Depending on the personal and compute time that you have for this, you could even get fancier and have filters for the average estimated reading time of the blog (eg. do you want a blog "snack" or a blog "meal") and its reading difficulty level?


https://www.pubmed.ai will generate a deep research report on that specific query involving checkpoint inhibitors and non small cell lung cancer for free. It won't have the nice comparison table that Elicit can generate. Also, pubmed.ai cannot parse a query as complex as the one used for Elicit, but you can paste the Elicit query into OpenAI's chatGPT and ask it to translate the query into a form that Pubmed can parse. Pubmed.ai was able to parse the chatGPT generated query string "("Checkpoint inhibitors"[Title/Abstract] OR "immune checkpoint blockade"[Title/Abstract])AND ("non-small cell lung cancer"[Title/Abstract] OR NSCLC[Title/Abstract])AND ("platinum-based chemotherapy"[Title/Abstract] OR cisplatin[Title/Abstract] OR carboplatin[Title/Abstract])AND (bevacizumab[Title/Abstract] OR "anti-VEGF"[Title/Abstract])AND ("PD-L1 expression"[Title/Abstract] OR "PD-L1 level"[Title/Abstract])AND (effectiveness[Title/Abstract] OR efficacy[Title/Abstract])AND (safety[Title/Abstract] OR toxicity[Title/Abstract])."




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: