Posting some theorems, and occasionally other stuff. By @bahran_cihan

Joined July 2022
Some theorems retweeted
Replying to @plain_simon
This might also stem from them having only 16 references which is absurd for a paper of this length
1
1
22
2,252
[Brehm–Kühnel 1987]: For fixed d ∈ℤ⁺, every combinatorial d-manifold with strictly less than 3⌈d/2⌉ + 3 vertices is a d-sphere.
6
3
55
Up to oriented diffeomorphism, - There is a unique smooth structure on the 61-sphere. - There are exactly 24 smooth structures on the 62-sphere. - There are exactly 142211872163171481167115958878208 smooth structures on the 63-sphere.
6
42
7
323
The "look and say sequence" {L[n]} goes 1, 11, 21, 1211, 111221, 312211, 13112221, ... [Conway 1986]: Writing d[n] for the number of digits in L[n], the limit of the ratio d[n+1]/d[n] exists, and is an algebraic integer of degree 71.
2
13
2
63
The quadratic integer ring ℤ[√14] is a Euclidean domain, but 𝐍𝐎𝐓 with respect to the natural norm N(a + b√14) := a² - 14b². The first unconditional proof of the Euclidean-ness of ℤ[√14] was by Harper in 2004.
4
45
The maximum number of colors necessary to properly color a map drawn on a closed connected surface with Euler characteristic χ is ⌊½(7 + sqrt(49-24χ))⌋, with a single exception: for the Klein bottle it is 6 (not 7). The 2-sphere case is the Four Color Theorem.
2
4
2
65
[Kirby–Siebenmann]. TOP/PL = K(ℤ/2, 3).
10
1
71
13,485
42 = (-80538738812075974)³ + (80435758145817515)³ + (12602123297335631)³
6
9
5
178
33,000
[Antonyan–Dobrowolski 2015]. For a locally compact Hausdorff group G with a closed subgroup H, TFAE: (1) G/H is locally contractible. (2) G/H is a manifold.
[Antonyan 2011]: For a compact Hausdorff group G with a closed subgroup H, TFAE: (1) The coset space G/H is contractible. (2) H = G.
10
74
8,233
Some theorems retweeted
1
7
1
97
7,772
[Bosboom–Demaine–Hesterberg–Lynch–Waingarten 2015]. It is PSPACE-hard to decide whether Player 1 has a forced win in a two-player Mario Kart race from given starting positions for the players.
2
35
4
209
21,597
Every position of Rubik's cube can be solved in at most 20 half-turn moves. Moreover there are positions for which 19 moves is not enough.
4
20
187
17,503
[Anderson 1977]. For a commutative ring R, every Artinian R-module is countably generated.
1
6
1
71
5,792
[Elkies–Klagsbrun 2024]. The elliptic curve y² + xy = x³ – 27006183241630922218434652145297453784768054621836357954737385x + 55258058551342376475736699591118191821521067032535079608372404779149413277716173425636721497 has rank ≥ 29.
3
11
3
188
11,560
[Cohen–Moore–Neisendorfer]: For every odd prime p, the least power of p that annihilates the p-primary part of all the homotopy groups of S²ⁿ⁺¹ is pⁿ.
2
16
The ℤ-rank of the homotopy groups of a simply-connected finite CW-complex of dimension d either become zero in degrees ≥ 2d, or they exhibit exponential growth.
3
5
53
[Basu–Basu 2015]: The homotopy groups of a simply-connected closed topological 4-manifold are completely determined by its second Betti number.
15
2
108
[Gorshkov 2013]. Given n≥5, TFAE: (1) Every finite group G with ω(G) = ω(A_n) is isomorphic to A_n. (2) n ∉ {6, 10}.
Given an element g of a group G, write |g| ∈ ℤ⁺ ⋃ {∞} for its order, and set ω(G) := {|g| : g∈G}. [Lytkina–Kuznetsov 2007]. Given a group G, TFAE: (1) ω(G) = {1, 2, 3, 4, 7}. (2) G ≅ PSL₂(7).
1
13
1
104
16,426
It is consistent with ZF that there is an 𝔽₂-vector space V with GL(V) = C₃. Note that GL(𝔽₂⊕𝔽₂) = S₃ .
5
8
5
77
[Tachtsis 2018]: ZF has a model in which the Boolean prime ideal theorem holds, but also there is a free group with a non-free subgroup.
5
1
54