Got a website?

Feel like helping make unauthorized LLM scrapers choke on an infinite sea of garbage, potentially making their models collapse?

...Then take a look at:
https://zadzmo.org/code/nepenthes/

PS Do make sure to read the warnings, boost and have fun!

.

Thanks to @dlatchx for reminding me where to find this!

I’ve created my own #Markov #Bot: RoboMOULE!

Inspired by others’ Markov / “ebooks” bots, @RoboMOULE will string words from 2,696 of my toots and toot in the style of me every 2 hours. Some of the results are already hilarious!

Thank you to @Erpel for making the Mame-Brot source code this bot uses: https://codeberg.org/fotoente/Mame-Brot

It’s my 5th bot! All my bots (@FediEverything, @ESEverything, @EverythingNow, @BAEverything, @RoboMOULE) will toot every 2 hours; 24 minutes apart from each other.

Been working on this project for a while, but think I'm ready to make it public. Hope to write more about it soon.

"The videos you are seeing are news clips from the day before today.

The text is generated from the subtitles for those videos, alongside a journal of mine from late October to early November 2016.

The video ordering is purely random, but the text follows its own internal logic. "

https://yesterdaysnews.today/

I don't remember much about #Markov Chains, #Random-Walk. Fokker-Planck. Jeez Louise. Scratched around trying to read and understand stuff. Fortunately at the time, my advisor cleared the decks, said don't bother, here is a 2-D #Gaussian dispersion model I wrote in BASIC. Re-write it in #FORTRAN and add two flavors: a #Lagrangian and a K-theory particle dispersion model, for a convective boundary layer. Use those formulations for a connective boundary layer you learned in Turbulence.

Re-posting because I'm still kind of in love with this project. A Markov-powered homage to Gertrude Stein's "Tender Buttons":

https://tenderbuttons.click/

i’ve built a daft “fake terminal” for my domain https://bleepbloop.net/ because I wanted to play with a #Markov chain text generator (a module which is trained from text you supply to it, and it uses this dataset to generate “predictive text”) and simulate what a malfunctioning AI assistant who’s circuits are frying might say as it is shut down. Think “Hal at the end of 2001 as the processors are being shut off”. I’ll probably continue tweaking it but it’s fun enough to share.

#MultiTarget tracking airborne #radar systems have been implemented using traditional #statistical techniques. Blackman's 1986 textbook is the leading publication on this subject.

Multi-target tracking is a sophisticated application of the fundamental idea behind the 1960s #Kalman filter—sequence prediction—which began with the 1900s #Markov chain. In the 2020s, #LLMs are perhaps the most appropriate technique for solving this age-old, tough problem.

RIP Dmitry #Markov

Gestern ist nicht nur Alexei #Navalny, sondern auch der russische Fotograf Dmitry Markov https://en.wikipedia.org/wiki/Dmitry_Markov_(photographer) gestorben. Auch er war Regimekritiker und zeitweilig im Gefängnis.

Er hat seine starken Fotos meistens nur mit einem iphone aufgenommen.

https://www.themoscowtimes.com/2024/02/16/russian-photographer-dmitry-markov-dies-at-42-a84121

Via @juliadavisnews Meanwhile in Russia: appearing on state TV, Igor #Markov asserted that #Russia must liquidate #Ukraine and the very notion of being Ukrainian has to be eradicated.

http://youtu.be/iSQvap630T0

Well, if you had told me you used *algorithms* in your code, I would never have doubted your finding of the "ultimate swear word according to math".

https://www.sciencefocus.com/news/heres-the-worlds-ultimate-swearword-according-to-maths/

#AI based #password candidate generation is super interesting to me, but so far the best generators (CMU's #neuralnetwork based generator presented at #USENIX 16, and Dr Matt Weir's #PCFG generator) are still about on par with existing #Markov implementations and rule-based #cracking.

#PassGAN, on the other hand, falls well short of existing password cracking techniques and candidate generators.

While I hold out a lot of hope for future AI-powered password cracking, it ain't here yet, and it definitely isn't PassGAN.

https://arstechnica.com/information-technology/2023/04/the-passgan-ai-password-cracker-what-it-is-and-why-its-mostly-hype/

English towns generated with a third-order #Markov chain:

Accrington, Alnwith-upon-in-Croseley, Amesbury, Bebingwood, Bexley and Ewell, Bish, Blackporth, Blaistol, Bognor Regis, Bognorton, Bostwood, Bradford, Brentree, Burford, Canterley, Chelsend, Cleckering Castbury, Crowston, Desbury, Edge, Faringwoodston & Lyness, Fleet Deephen, Frith, Frith-Wolsingtone, Gard, Gill, Grassett, Hast Town, Heley, Hunthampton, Kempstone, Kenistlesbury Crowbridge, Keynemouth, Kingdon, Leominster, Letchesham, Lich, Loddock, Ludge, Ludley, Luton, Lutone Torrington Stratford, Made, Maghull, Mers, Midhurston, Minster, More, Newlyn, Newmarkelton, Newton, Oakham Abbot, Orpington, Pick, Purleethe-Sea, Radston, Reigh, Royal Leaminstantefract, Rugby, Scar, Settesford, Shield, Sleach witherhildeanorton Latimerehampston, South, Thorlandham, Thorn, Thorpe andwickham Fernstabley, Wade, Waltenstanhoe, Wanthorpenny St Merston, Watch, Wester and Birksop, Wood, Woodbridlington, Wooler, Yatell