This is one of those jokes you make, then realize people in the field have probably been saying it verbatim for 20 years.
COMIC ◆ https://t.co/Hic22AtO7Z
PATREON ◆ https://t.co/DYYMqy2KN8
STORE ◆ https://t.co/aRRj5Lx651
A Black Hole’s Puzzling X-Ray Bursts
A supermassive black hole awakened seven years ago, and it’s now behaving in a way astronomers can’t explain. https://t.co/XOM4k8JFeT
Years ago when my wife and I we were planning to buy a home, my dad stunned me with a quick mental calculation of loan payments.
I asked him how -he said he'd learned the strange formula for compound interest from his father, who was a merchant born in 19th century Iran
1/4
STEM is so cool. Science is amazing. Kids should develop curiosity about how the universe works and proficiency in uncovering its secrets, so that they can most effectively optimize clickthrough rates on banner ads at the bottom of VC-funded website-apps.
Our NICER telescope on the @Space_Station is studying mysterious events called quasi-periodic eruptions, or QPEs. These repeating X-ray flares happen near the central black holes in some galaxies. QPEs in a system nicknamed Ansky are the most energetic and longest yet discovered! https://t.co/cedv2CavoD
So where are all the outspiraling double WDs? Maybe they all merge, maybe we’re not detecting them, or something else altogether! For more, including eerily identical optical spectra and further discussion of evolutionary models, check out the paper https://t.co/6GweSzPIAR (6/6)
Belated paper day: https://t.co/6GweSzPIAR
We discovered three ultracompact binaries with accreting white dwarfs (WDs), by their photometric and spectroscopic variability. With orbital periods of 13.2/8.7/7.9 mins, they are the most compact binaries with accretion disks! (1/6)
Finally, there’s emerging tension b/w theory & data. MESA models of ELM+CO WDs predict the initial inspiral/period minimum should be >>faster than the subsequent outspiral. We should blindly detect |Pdot_in/Pdot_out|~4x more outgoing systems at 10 mins. That is… not seen. (5/6)
🚨 What’s the best way to select data for fine-tuning LLMs effectively?
📢Introducing ZIP-FIT—a compression-based data selection framework that outperforms leading baselines, achieving up to 85% faster convergence in cross-entropy loss, and selects data up to 65% faster.
🧵1/8