Alphabeta Math
Session-authored (Fable 5 assisted)
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced — the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted — a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated — a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

8 results · all verified · 0 also independently AI-judged
Every result on this page is machine-checked by a proof checker and read in full and owner-audited; the judge is an additional, independent cross-model AI review of the proofs. The 8 not AI-judged were verified by owner audit (typically over a confirmed judge false positive), not failures.

Chains, Antichains, Sperner and Dilworth — Examples

1 · Prerequisites

2 · Summary

3 · Logical flowchart

4 · Definitions, theorems and proofs

None yet.

5 · Examples, counterexamples and false statements

ExampleConstruction: AI-adaptedVerification: AI-generatedprecheck passaudited 2026-07-31Open item page →

The Boolean lattice on four elements: ranks, width, shadows, and a symmetric chain decomposition

Example

Let A={1,2,3,4}A=\{1,2,3,4\}. The Boolean lattice B(A)B(A) has rank sizes

1, 4, 6, 4, 1,1,\ 4,\ 6,\ 4,\ 1,

so its width is 66. For F={{1,2},{1,3}}\mathcal F=\{\{1,2\},\{1,3\}\}, both its lower and upper shadows have three members:

F={{1},{2},{3}},F={{1,2,3},{1,2,4},{1,3,4}}.\partial\mathcal F=\{\{1\},\{2\},\{3\}\},\qquad \nabla\mathcal F=\{\{1,2,3\},\{1,2,4\},\{1,3,4\}\}.

;12341213142324341231241342341234B(A)0B(A)1B(A)2B(A)3B(A)4@FFrF
middlerankrank0rank1rank2rank3rank4C1;1121231234C2414124C3223234C424C5313134C634

Facts & Assumptions

Given: The set A={1,2,3,4}A=\{1,2,3,4\} and the family F\mathcal F in the Example.

[F1]

The rank-kk level of B(A)B(A) is the family of kk-subsets (The Boolean lattice of subsets of a finite set and its rank levels).

[F2]

Lower and upper shadows consist of the immediate subsets and supersets one rank away (The lower and upper shadows of a uniform set family).

[L1]

Sperner's theorem says the width of B(A)B(A) is its middle binomial coefficient (Sperner's theorem and its equality cases: a largest antichain is a complete middle level).

[L2]

Every finite Boolean lattice has a symmetric chain decomposition (Every finite Boolean lattice has a symmetric chain decomposition).

Verification

technique · direct
1.1

Listing subsets by cardinality gives rank sizes 1,4,6,4,11,4,6,4,1, and [L1] gives width 66.

givenF1L1
1.2

Deleting one element from a member of F\mathcal F gives exactly {1},{2},{3}\{1\},\{2\},\{3\}, while adjoining one element gives exactly 123,124,134123,124,134. Thus the displayed shadows are correct.

givenF2
1.3

The following symmetric chains partition all sixteen subsets: 1121231234\varnothing\subset1\subset12\subset123\subset1234; 4141244\subset14\subset124; 2232342\subset23\subset234; 2424; 3131343\subset13\subset134; and 3434. Their endpoint ranks sum to 44, in agreement with [L2].

F1L2
2.1

Steps 1.1, 1.2, and 1.3 verify the ranks, width, shadows, and an explicit symmetric chain decomposition.

step 1.1step 1.2step 1.3
ExampleConstruction: AI-generatedVerification: AI-generatedprecheck passaudited 2026-07-31Open item page →

A six-element poset of width three and a three-chain cover

Example

Let P={a1,a2,a3,b1,b2,b3}P=\{a_1,a_2,a_3,b_1,b_2,b_3\} with ai<bia_i<b_i for each ii, and no other comparabilities between distinct elements. Then {a1,a2,a3}\{a_1,a_2,a_3\} is an antichain, and

{a1<b1},{a2<b2},{a3<b3}\{a_1<b_1\},\qquad\{a_2<b_2\},\qquad\{a_3<b_3\}

is a chain cover. The width is exactly 33, and this cover is minimum.

a1a2a3b1b2b3C1C2C3maximumantichainfa1;a2;a3g

Facts & Assumptions

Given: The six-element poset PP described in the Example.

[L1]

In a finite poset, the minimum number of chains in a chain cover equals the width (Dilworth's theorem: the minimum number of chains covering a finite poset equals its width).

Verification

technique · direct
1.1

The set {a1,a2,a3}\{a_1,a_2,a_3\} is an antichain, so the width is at least 33.

given
1.2

Every antichain contains at most one element from each comparable pair {ai,bi}\{a_i,b_i\}, so it has at most 33 elements. Hence the width is exactly 33.

given
1.3

The three displayed two-element chains cover all six elements, so they form a chain cover of cardinality 33.

given
2.1

By steps 1.2 and 1.3, and equivalently by [L1], the displayed cover has the minimum possible number of chains.

step 1.2step 1.3L1
ExampleConstruction: AI-adaptedVerification: AI-generatedprecheck passaudited 2026-07-31Open item page →

The divisors of 6060 form a finite distributive lattice and realize Birkhoff's representation concretely

Example

Order the positive divisors of 6060 by divisibility. Since 60=223560=2^2\cdot3\cdot5, every divisor has a unique form

2α3β5γ,0α2,0β,γ1.2^\alpha3^\beta5^\gamma,\qquad 0\le\alpha\le2,\quad0\le\beta,\gamma\le1.

Divisibility is componentwise comparison of the exponent triples. Meet and join are componentwise minimum and maximum, so this is a finite distributive lattice.

135152610304122060join-irreducibledÁewhene=disprime

Facts & Assumptions

Verification

technique · direct
1.1

By [L1], the exponent-triple description is unique, and ded\mid e exactly when every exponent of dd is at most the corresponding exponent of ee.

givenL1
2.1

Componentwise minimum and maximum give the greatest common divisor and least common multiple, and the distributive identities hold coordinatewise for minimum and maximum on chains. Thus the divisor poset is a finite distributive lattice.

step 1.1algebra
2.2

Its join-irreducibles are 2,4,3,52,4,3,5. In their inherited order, 2<42<4 and 3,53,5 are incomparable with these and with each other.

step 1.1
3.1

A divisor dd maps to the order ideal of join-irreducibles dividing it: its 22-exponent chooses \varnothing, {2}\{2\}, or {2,4}\{2,4\}, while its 33- and 55-exponents independently choose whether to include 33 and 55. This is exactly the Birkhoff map of [L2].

step 2.2L2
4.1

Hence the divisors of 6060 concretely realize Birkhoff's representation as the order ideals of the poset 2<42<4 with isolated elements 33 and 55.

step 2.1step 3.1
ExampleConstruction: AI-generatedVerification: AI-generatedprecheck passaudited 2026-07-31Open item page →

Four explicit petals with a common two-element core form a sunflower

Example

The four sets

{1,2,3,4},{1,2,5,6},{1,2,7,8},{1,2,9,10}\{1,2,3,4\},\quad\{1,2,5,6\},\quad\{1,2,7,8\},\quad\{1,2,9,10\}

form a 44-petal sunflower. Their common core is {1,2}\{1,2\} and their petals are the pairwise disjoint sets {3,4}\{3,4\}, {5,6}\{5,6\}, {7,8}\{7,8\}, and {9,10}\{9,10\}.

f1;2gf3;4gf5;6gf7;8gf9;10gF1F2F3F4commoncoreFi=f1;2g[Pi,withthePipairwisedisjoint

Facts & Assumptions

Given: The four sets displayed in the Example.

[F1]

Distinct sets form a sunflower when all pairwise intersections equal one common core (Sunflowers, petals, and their common core).

Verification

technique · direct
1.1

Every displayed set contains {1,2}\{1,2\}, and outside this pair their elements lie in disjoint two-element blocks.

given
2.1

Therefore the intersection of any two distinct displayed sets is exactly {1,2}\{1,2\}.

step 1.1
3.1

By [F1], the four sets form a sunflower with the stated core and petals.

step 2.1F1
ExampleConstruction: AI-adaptedVerification: AI-generatedprecheck passaudited 2026-07-31Open item page →

All kk-sets through a fixed point form an intersecting family attaining the Erdős-Ko-Rado bound

Example

Let AA be an nn-element set with 1k1\le k and n2kn\ge2k, and fix aAa\in A. The star

Sa={S[A]k:aS}\mathcal S_a=\{S\in[A]^k:a\in S\}

is intersecting and has cardinality (n1k1)\binom{n-1}{k-1}, attaining the Erdős-Ko-Rado bound.

Facts & Assumptions

Given: An nn-element set AA, natural numbers 1k1\le k and n2kn\ge2k, and a point aAa\in A.

[L1]

Erdős-Ko-Rado bounds an intersecting family of kk-subsets by (n1k1)\binom{n-1}{k-1} and states that a star attains the bound (Erdős-Ko-Rado theorem: for 1k1\le k and n2kn\ge 2k, an intersecting family of kk-subsets of an nn-set has size at most (n1k1)\binom{n-1}{k-1}, and a star attains the bound).

Verification

technique · direct
1.1

Any two members of Sa\mathcal S_a intersect at aa, so the star is intersecting.

given
1.2

The map SS{a}S\mapsto S\setminus\{a\} is a bijection from Sa\mathcal S_a to the (k1)(k-1)-subsets of A{a}A\setminus\{a\}. Hence Sa=(n1k1)|\mathcal S_a|=\binom{n-1}{k-1}.

givenF1
2.1

By [L1], step 1.2 equals the universal upper bound, so the star is extremal.

step 1.1step 1.2L1
CounterexampleConstruction: AI-adaptedVerification: AI-generatedprecheck passaudited 2026-07-31Open item page →

A maximal antichain of size one in a finite poset of width two

Statement refuted

The false statement False: every maximal antichain in a finite poset has maximum cardinality claims that every maximal antichain in a finite poset has maximum cardinality.

Facts & Assumptions

Given: The poset P={a,b,c}P=\{a,b,c\} with a<ba<b, a<ca<c, and b,cb,c incomparable.

[F1]

An antichain is maximal when no larger antichain contains it, and maximum when no antichain has greater cardinality; the width is the maximum cardinality of an antichain (Antichains, chain covers, and antichain covers of a poset, Height and width of a nonempty finite poset).

Counterexample

technique · direct
1.1

The singleton {a}\{a\} is an antichain and cannot be enlarged, since both bb and cc are comparable with aa.

givenF1
1.2

The pair {b,c}\{b,c\} is an antichain. No three-element antichain exists, because the only three-element subset is PP itself and it contains the comparable pair a,ba,b; hence the width of PP is 22.

givenF1
2.1

Thus {a}\{a\} is maximal of cardinality 11 but not maximum, providing the required counterexample.

step 1.1step 1.2

Remarks

abcmaximalantichainfagmaximumantichainfb;cg
CounterexampleConstruction: AI-adaptedVerification: AI-generatedprecheck passaudited 2026-07-31Open item page →

When k<n<2kk<n<2k, the entire kkth level is intersecting and exceeds the Erdős-Ko-Rado star bound

Statement refuted

The false statement False: the Erdős-Ko-Rado bound holds without the hypothesis n2kn\ge 2k claims the Erdős-Ko-Rado star bound without assuming n2kn\ge2k.

Facts & Assumptions

Given: Natural numbers satisfying exactly k<n<2kk<n<2k, an nn-element set AA, and the full level F=[A]k\mathcal F=[A]^k.

[F1]

An intersecting family has nonempty intersection between every pair of members, and (nk)=[A]k\binom nk=|[A]^k| (Intersecting uniform families of finite sets, The set [A]k[A]^{k} of kk-element subsets and the binomial coefficient (nk):=[n]k\binom{n}{k} := \lvert [n]^{k}\rvert).

Counterexample

technique · direct
1.1

If S,T[A]kS,T\in[A]^k were disjoint, then ST=2k>n=A|S\cup T|=2k>n=|A|, impossible. Hence the entire level F\mathcal F is intersecting.

givenF1
1.2

By [L1], F=(nk)=(n/k)(n1k1)|\mathcal F|=\binom nk=(n/k)\binom{n-1}{k-1}. Since k<nk<n, the factor n/kn/k is greater than 11, so F>(n1k1)|\mathcal F|>\binom{n-1}{k-1}.

givenF1L1algebra
2.1

Thus for every k<n<2kk<n<2k, the full kkth level is an intersecting family larger than a star, refuting the bound outside its stated range.

step 1.1step 1.2
CounterexampleConstruction: AI-adaptedVerification: AI-generatedprecheck passaudited 2026-07-31Open item page →

The diamond M3M_3 and pentagon N5N_5 violate distributivity by explicit joins and meets

Statement refuted

Every finite lattice is distributive.

Facts & Assumptions

Given: The diamond M3={0,1,a,b,c}M_3=\{0,1,a,b,c\}, where a,b,ca,b,c are incomparable atoms, and the pentagon N5={0,a,b,c,1}N_5=\{0,a,b,c,1\}, where 0<a<b<10<a<b<1, 0<c<10<c<1, and cc is incomparable with a,ba,b.

[F1]

Distributivity requires x(yz)=(xy)(xz)x\wedge(y\vee z)=(x\wedge y)\vee(x\wedge z) for all elements (Lattices, distributive lattices, and order ideals).

Counterexample

technique · direct
1.1

In M3M_3, one has bc=1b\vee c=1, ab=0a\wedge b=0, and ac=0a\wedge c=0. Therefore a(bc)=aa\wedge(b\vee c)=a, while (ab)(ac)=0(a\wedge b)\vee(a\wedge c)=0.

givenF1
1.2

In N5N_5, one has ac=1a\vee c=1, ba=ab\wedge a=a, and bc=0b\wedge c=0. Therefore b(ac)=bb\wedge(a\vee c)=b, while (ba)(bc)=a(b\wedge a)\vee(b\wedge c)=a.

givenF1
2.1

Since a0a\ne0 in M3M_3 and aba\ne b in N5N_5, each lattice violates the distributive identity in [F1]. Both are finite, so either one refutes the Statement.

step 1.1step 1.2F1

Remarks

M30abc1N50abc1

Sources