mfscholar

For your reference, check the 3M map of science papers we made here: https://www.scholar-inbox.com/scholar-maps

If you're looking to set up a habit cycle, I'd recommend three steps:

1. Find a cue that will remind you to start writing, e.g. having your morning coffee

2. Write any amount of time; say 30min or so

3. Reward yourself. I just have a little snack, but it could be anything

Works great for me, and I found once I changed some small habits, it was also easier to do better overall. This advice is from the book "The power of habit" by Charles Dhuigg

Embedding size and maximum context length are not related. The maximum context length of gpt-3.5-turbo is known though: 2^14

We just released Scholar Inbox as part of my PhD.

It is a research paper recommendation system, which downloads the daily arXiv.org research papers and then makes personalised recommendations to the user.

You name some authors that you like, including yourself, and upvote their most interesting papers. Then the system starts recommending papers to you. You can +/- vote papers which further trains the system. Finally, you choose how often you would like to receive a summary email, listing your top recommendations. If you don't want the summary email, you can just visit the site and it will aggregate and rank all papers that you missed since your last visit.

My supervisor made this post on X: https://twitter.com/AutoVisionGroup/status/17375715672849451...

Why not buy her a gold bar instead? They can be as expensive/ valuable as you like and she can see it as an investment. It can't be worn on a finger of course, but where I'm from, people only wear the engagement ring before their wedding.

Continual Learning, a ML training concept