Dmitry Rybin disproves Dinitz-Garg-Goemans conjecture using GPT-5.6 Pro

So ppl are gonna really just continue this trend of resolving decade old conjecture within a tweet and chatgpt link? https://twitter.com/DmitryRybin1/status/2079904005652893709

likes: 1.8K
replies: 49
bookmarks: 240
reposts: 46

EB

Erik Bernhardsson@bernhardsson 5 weeks ago

Going to tell my wife our date night is cancelled since I need to feed every open conjecture into ChatGPT asking for counterexamples. https://twitter.com/dmitryrybin1/status/2079904005652893709

likes: 403
replies: 18
bookmarks: 54
reposts: 10

FF
François Fleuret@francoisfleuret 5 weeks ago
@bernhardsson Bro ask the thing to write stuff to ask itself the things, and have sushis with the wife.
- likes: 8
- replies: 1
- bookmarks: 0
- reposts: 0
PD
Pim de Witte@PimDeWitte 5 weeks ago
@bernhardsson How do I invest early in this banger
- likes: 2
- replies: 0
- bookmarks: 0
- reposts: 0

B(

Beff (e/acc)@beffjezos 5 weeks ago

Another one bites the dust. AI is like a hot knife cutting through mathematical butter. No conjecture is safe anymore. https://twitter.com/DmitryRybin1/status/2079904005652893709

likes: 514
replies: 27
bookmarks: 65
reposts: 22

R@

rishi @ ICML@RishiBommasani 5 weeks ago

For humans, persistence often is only possible given signs of progress. In the quest for counterexamples, mathematicians may quit too early because they cannot see these signs. For machines, such incremental feedback may not be necessary. That seems like a big deal.…

likes: 11
replies: 0
bookmarks: 3
reposts: 0

Dmitry Rybin disproves Dinitz-Garg-Goemans conjecture using GPT-5.6 Pro

Combined views

Dmitry Rybin disproves Dinitz-Garg-Goemans conjecture using GPT-5.6 Pro

Reactions

Combined views