← Feed

Posts

• View post

  Today marks the first paper together with co-authors from TU Delft. More than four years after I first joined that university, and more than one year after leaving it. This is symptomatic of the omnipresent short-term contracts in the academic world, and the resulting need to "start over" every time. On the one hand you get to see many perspectives. On the other hand the calibration costs are comparitively high and by the time you become productive, you have to leave already. A...

• View post

  The hunt for disjoint list colorings in planar graphs continues, this time together with Daniel Cranston and František Kardoš: https://arxiv.org/abs/2602.16692 Is it possible to find two disjoint L-colorings for every 5-fold list-assignment L? If so, it implies three disjoint L-colorings for every 6-fold L. We proved the latter unconditionally. (More related results on K5-minor-free graphs are in an earlier work together with Cambie and Zhu: https://arxiv.org/abs/2312.17233 )

• View post

  What to do if you get a review request for a paper that deserves to be published in a much much higher journal? As a reviewer it seems you only have some influence to send a paper "down", but never to correct upwards.

• View post

  RE: https://mathstodon.xyz/@skelk/116174927025037600 Now with an AI-assisted and LEAN formalized proof, available at https://woutercvb.github.io/greedy_by_finish_time_draft22april2026.pdf

• View post

  Ramsey problem solved by some internal model at OpenAI: for all k, the ratio R(k,x+1)/R(k,x) converges to 1, as x goes to infinity. https://cdn.openai.com/pdf/6dc7175d-d9e7-4b8d-96b8-48fe5798cd5b/Ramsey.pdf Crazy times.

• View post

  Getting close to completing my collection of combinatorics journals! (Just in time before AI takes over our profession :S) https://www.sciencedirect.com/science/article/pii/S0095895626000250

• View post

  The Erdös unit distance conjecture has apparently been disproved by an internal model of OpenAI (or is it ClosedAI then?). Impressive stuff. Guess it's only a matter of time now before my own favourite problems get absorbed and conquered as well. So far none of them have been touched, but maybe next year all our papers will be obsolete? Sometimes I wonder how many failed AI attacks there have been on a particular problem; only the positive outcomes are reported.