Sorry I can't give out any bsky invites. I need to invite my 99 other alt accounts first
30.04.2023 23:07 — 👍 5 🔁 0 💬 0 📌 0@einschtein.bsky.social
2D crypto data-dog; IRL applied AI/ML/NLP in academic medicine; proud 1x winner of the Bloomberg Podcast What Goes Up's Craziest Things in Markets This Week
Sorry I can't give out any bsky invites. I need to invite my 99 other alt accounts first
30.04.2023 23:07 — 👍 5 🔁 0 💬 0 📌 0I totally think sbert is a great place to start out of the box for good embeddings without needing to fine tune. Throw on a linear classifier on top (something like sklearn logistic regression to keep it simple) and IMO you could get pretty far on something fairly lightweight and fast.
26.04.2023 10:02 — 👍 5 🔁 1 💬 2 📌 0One of the best parts about chatgpt is it's massive context window (easily in the 4000-8000 token range for gpt3.5). Many sbert models top out at a window of 512 tokens or roughly 300 words on average. Not too bad if you're only classifying a post/reply on its own
26.04.2023 10:10 — 👍 4 🔁 0 💬 0 📌 0Huggingface may have some decent models too but they may be overly optimized for certain datasets (ie sms spam)
https://huggingface.co/models?search=spam
Def not going to be the best but should be an informative baseline. Starting out you could aim to filter out the most extreme stuff and focus on optimizing high precision so that only the most obvious spam gets filtered out (ie fake crypto scam giveaways).
Assumes that you have training data tho
I totally think sbert is a great place to start out of the box for good embeddings without needing to fine tune. Throw on a linear classifier on top (something like sklearn logistic regression to keep it simple) and IMO you could get pretty far on something fairly lightweight and fast.
26.04.2023 10:02 — 👍 5 🔁 1 💬 2 📌 0Some of this volume is attributed to paper mills - - these are groups that sell "ready to publish" research papers to academics to inflate a career advancing metric of published articles. Unfortunately, many paper mill research papers are often falsified / use made up data
25.04.2023 03:54 — 👍 0 🔁 0 💬 0 📌 0Several "mega journals" (ie IJERPHealth which published 17000 papers in 2022) are getting delisted due to concerns around lack of quality control / peer review.
https://www.chemistryworld.com/news/sanctioning-of-50-journals-raises-concerns-over-special-issues-in-mega-journals/4017315.article
Inaugural session on SkySpaces.
SkySpaces is a community-built platform for audio social features on BlueSky by @geeken.tv 🙌
(ELI5: Twitter Spaces for BlueSky)
Check it out at https://skyspaces.net/
Wonder if we'll see Vitalik come over now that there's an Android app.
I heard that he liked Poaster but was sad it was iOS only
So real
What's really upsetting is seeing the book take off and lend them cred
I've seen this in medicine. Curious about what fields you're looking at -- would love to know if this is a universal acaremic thing
Yeah, we are somehow descending / discovering the lowest common denominator version of many ideas / ideologies
14.04.2023 12:59 — 👍 2 🔁 0 💬 1 📌 0LA sure but I'm not confident about New England drivers haha
12.04.2023 18:13 — 👍 1 🔁 0 💬 0 📌 0👀
12.04.2023 05:21 — 👍 0 🔁 0 💬 0 📌 0* farcaster
12.04.2023 05:20 — 👍 1 🔁 0 💬 0 📌 0we simply vibe
either you make it or you don't. might as well ride the wave
not denying that there will be societal upheaval, but I believe the cat is out of the bag
gm
trying out one more social app and community after: warpcaster, lens, poaster, and ofc crypto twitter