HackerTrans
TopNewTrendsCommentsPastAskShowJobs

mpmisko

no profile record

Submissions

World Models for Planning Agents

mpmisko.github.io
2 points·by mpmisko·2 bulan yang lalu·0 comments

Show HN: MediSearch Pro–most accurate medical question-answering system

medisearch.io
1 points·by mpmisko·2 tahun yang lalu·0 comments

Show HN: Verify medical claims in TikToks and YouTube Shorts

medisearch.io
1 points·by mpmisko·2 tahun yang lalu·1 comments

Fun times with energy-based models

mpmisko.github.io
82 points·by mpmisko·2 tahun yang lalu·14 comments

AI Fundamentals: Energy-Based Models

mpmisko.github.io
2 points·by mpmisko·2 tahun yang lalu·0 comments

Solving Death with AI? A Deep Dive into Energy-Based Models

mpmisko.github.io
1 points·by mpmisko·2 tahun yang lalu·0 comments

AI Fundamentals: Energy-Based Models

mpmisko.github.io
5 points·by mpmisko·2 tahun yang lalu·0 comments

What happened to blogs

mpmisko.github.io
70 points·by mpmisko·2 tahun yang lalu·61 comments

[untitled]

2 points·by mpmisko·2 tahun yang lalu·0 comments

Show HN: I put PubMed in a vector DB

pubmedisearch.com
97 points·by mpmisko·2 tahun yang lalu·27 comments

Show HN: Medical LLM API on par with Google Med-PaLM 2. 92% USMLE accuracy

medisearch.io
17 points·by mpmisko·3 tahun yang lalu·13 comments

comments

mpmisko
·2 tahun yang lalu·discuss
EBMs show up all over the place, apparently even your classifier is an EBM :) (https://arxiv.org/abs/1912.03263).
mpmisko
·2 tahun yang lalu·discuss
GPT-based search engines usually use some sort of a database to retrieve context for the LLM to summarize first. This is what people refer to as RAG these days: https://blogs.nvidia.com/blog/what-is-retrieval-augmented-ge....

Some of these GPT engines maintain their own vector DB to do semantic search, others are directly hooked into Bing / Google. So pubmedisearch.com would be one component of a GPT-based engine. We actually have a GPT-based engine here: https://medisearch.io/.
mpmisko
·2 tahun yang lalu·discuss
Will definitely check pgvector, thanks for the pointer.
mpmisko
·2 tahun yang lalu·discuss
Lots of annoying edge cases as you can imagine, nothing particularly glamorous.
mpmisko
·2 tahun yang lalu·discuss
Done! Let me know if you have other feedback.
mpmisko
·2 tahun yang lalu·discuss
Thanks! Looks quite relevant
mpmisko
·2 tahun yang lalu·discuss
Training for multiple epochs is a bit like that :)
mpmisko
·2 tahun yang lalu·discuss
We use pinecone and it is not ideal, looking at https://turbopuffer.com/ now. They look quite promising :)
mpmisko
·2 tahun yang lalu·discuss
1. We cover all the articles on PMC. The exact cost is hard to estimate because we did a lot of iterations.

2. We do weight those ... it is a lot of trial and error and you have to have good & exhaustive benchmarks.
mpmisko
·2 tahun yang lalu·discuss
Yes
mpmisko
·2 tahun yang lalu·discuss
Glad you like it! I did this as a mini-project within our startup MediSearch (https://medisearch.io/) & the search pipeline is custom tuned for the problem.
mpmisko
·2 tahun yang lalu·discuss
Hi, it currently does not support search by PMID. But you can find the paper included in the results here (5th place):

https://pubmedisearch.com/share/Do%20some%20individuals%20wi...
mpmisko
·2 tahun yang lalu·discuss
Just looking at stuff like citations and impact factors of journals.
mpmisko
·2 tahun yang lalu·discuss
Thanks!

It uses a vector search approach. Your query is embedded in a vector space using a language model and we find the closest vector to the query from the PubMed papers. This is a good summary of the techniques: https://learn.microsoft.com/en-us/azure/search/vector-search.... There are a couple more tricks but this is the gist.

The nice part is that this approach allows you to find relevant papers to your question. E.g, you can ask "Can secondhand smoke cause AMD?" and the very first few papers are answering your question (https://pubmedisearch.com/share/Can%20secondhand%20smoke%20c...). The more specific question, the better. :)
mpmisko
·3 tahun yang lalu·discuss
Thanks for giving it a go! We will charge for the API and we may have a pro subscription with some nice features :).
mpmisko
·3 tahun yang lalu·discuss
We primarily cite medical scientific papers and health guidelines. We occasionally cite health blogs (such as healthline.com), but this is mostly for simple health questions (e.g., "tell me some nutrition tips").
mpmisko
·3 tahun yang lalu·discuss
Hey, a React client is definitely on our roadmap! The demo (https://tinyurl.com/medisearch-docs) in our docs is written in JS, so it should be useful out of the gate! A proper client is coming in the next few weeks.
mpmisko
·3 tahun yang lalu·discuss
Glad that you like it! We do support follow-up questions. Implementing a chat-style interaction should be very straightforward :) (see our Python client for examples https://pypi.org/project/medisearch-client/).