Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Eqs. (6)–(10) — four-valent bases and recoupling

Proved
QuantumSixJ.Network.four_valent_bases

by igor.kh · Nov 13, 2026 · Mathlib 0df444a (Lean v4.33.1)

monoidal-categoriesquantum-groupsspin-networks

Let C\mathcal CC be a KKK-linear preadditive monoidal category, K=C(q)K=\mathbb C(q)K=C(q), with additive and KKK-linear tensor product. The chosen objects VnV_nVn​ for natural labels satisfy V0≅1V_0\cong\mathbb1V0​≅1 and have exact self-pairings. Assume SuQ2Axioms: unique scalar endomorphisms of each VnV_nVn​, zero morphisms between different labels, the admissible SL(2)\mathrm{SL}(2)SL(2) fusion decompositions, and the loop values Δn\Delta_nΔn​. Choose trivalent vertices WWW, and assume VertexNormalized W: theta normalization and cyclicity. Scalars act through ι:Q(q)↪K\iota:\mathbb Q(q)\hookrightarrow Kι:Q(q)↪K, fixing qqq. Composition f;gf;gf;g means first fff, then ggg.

Without TetSeed, prove the complete four-valent basis and recoupling statement for every natural a,b,c,da,b,c,da,b,c,d:

  1. Hk;HjR=δkjι(hj)id1H_k;H_j^R=\delta_{kj}\iota(h_j)\mathrm{id}_{\mathbb1}Hk​;HjR​=δkj​ι(hj​)id1​ and Xm;XlR=δmlι(xl)id1X_m;X_l^R=\delta_{ml}\iota(x_l)\mathrm{id}_{\mathbb1}Xm​;XlR​=δml​ι(xl​)id1​, including inadmissible channels.
  2. Every state f:1→Va⊗Vb⊗Vc⊗Vdf:\mathbb1\to V_a\otimes V_b\otimes V_c\otimes V_df:1→Va​⊗Vb​⊗Vc​⊗Vd​ has scalar pairing coefficients in each basis and equals its corresponding orthogonal expansion. Every covector has the dual expansion as well. Horizontal sums run over 0≤j≤b+c0\le j\le b+c0≤j≤b+c and vertical sums over 0≤l≤a+b0\le l\le a+b0≤l≤a+b; coefficients are divided by the respective norms.
  3. Prove the diagram identification
X(a,b,c,d;l);HR(a,b,c,d;j)=T(a,b,c,d;j,l),X(a,b,c,d;l);H^R(a,b,c,d;j)=T(a,b,c,d;j,l),X(a,b,c,d;l);HR(a,b,c,d;j)=T(a,b,c,d;j,l),

where the right-hand side is the explicitly defined graphical network X(a,d,c,b;j);HR(a,d,c,b;l)X(a,d,c,b;j);H^R(a,d,c,b;l)X(a,d,c,b;j);HR(a,d,c,b;l). This equality is a conclusion to establish, not an extra hypothesis or an assumed change of notation. 4. For every lll, scalar coefficients sjs_jsj​ exist with T(a,b,c,d;j,l)=sjid1T(a,b,c,d;j,l)=s_j\mathrm{id}_{\mathbb1}T(a,b,c,d;j,l)=sj​id1​ and

Xl=∑j=0b+csjι(hj)Hj.X_l=\sum_{j=0}^{b+c}\frac{s_j}{\iota(h_j)}H_j.Xl​=j=0∑b+c​ι(hj​)sj​​Hj​.

The statement includes coefficient existence rather than presupposing that all closures have already been evaluated. Division by zero follows the specified zero-extension convention.

Independent mathematical read-back

Four-valent pairings and expansions

Let K=C(q)K=\mathbb C(q)K=C(q), and let ι:Q(q)→K\iota:\mathbb Q(q)\to Kι:Q(q)→K be the coefficient embedding fixing qqq. Work in an arbitrary preadditive KKK-linear monoidal category with tensor product additive and KKK-linear in both variables. Write 1\mathbf11 for the tensor unit. Composition f≫gf\gg gf≫g means first fff, then ggg; canonical associators and unitors are implicit, never permutations of tensor factors.

Hypotheses and notation

For each natural aaa, an object VaV_aVa​ is given, together with V0≅1V_0\cong\mathbf1V0​≅1 and a self-duality cup uau_aua​ and cap eae_aea​ satisfying both snake identities. There is a chosen covariant vertex vabc:Va⊗Vb⊗Vc→1v_{abc}:V_a\otimes V_b\otimes V_c\to\mathbf1vabc​:Va​⊗Vb​⊗Vc​→1 for every natural triple.

A triple (a,b,c)(a,b,c)(a,b,c) is admissible when its entries are nonnegative, satisfy all three triangle inequalities, and have even sum. The assumed category axioms are: every endomorphism of VaV_aVa​ is a unique scalar multiple of its identity; all morphisms between distinctly labelled objects vanish; for every a,ba,ba,b there exist maps ic:Vc→Va⊗Vbi_c:V_c\to V_a\otimes V_bic​:Vc​→Va​⊗Vb​ and pc:Va⊗Vb→Vcp_c:V_a\otimes V_b\to V_cpc​:Va​⊗Vb​→Vc​, for all natural ccc, with ic≫pc=id⁡Vci_c\gg p_c=\operatorname{id}_{V_c}ic​≫pc​=idVc​​ for admissible c≤a+bc\le a+bc≤a+b, and ∑c≤a+b, (a,b,c) admissiblepc≫ic=id⁡Va⊗Vb\sum_{c\le a+b,\ (a,b,c)\text{ admissible}}p_c\gg i_c=\operatorname{id}_{V_a\otimes V_b}∑c≤a+b, (a,b,c) admissible​pc​≫ic​=idVa​⊗Vb​​; and ua≫ea=ι(Δ(a))id⁡1u_a\gg e_a=\iota(\Delta(a))\operatorname{id}_{\mathbf1}ua​≫ea​=ι(Δ(a))id1​.

Set [n]=(qn−q−n)/(q−q−1)[n]=(q^n-q^{-n})/(q-q^{-1})[n]=(qn−q−n)/(q−q−1), with [n]!=0[n]!=0[n]!=0 for negative integer nnn, and [n]!=∏k=1n[k][n]!=\prod_{k=1}^n[k][n]!=∏k=1n​[k] otherwise. For natural aaa, Δ(a)=(−1)a[a+1]\Delta(a)=(-1)^a[a+1]Δ(a)=(−1)a[a+1]. For admissible (a,b,c)(a,b,c)(a,b,c), put s=(a+b+c)/2s=(a+b+c)/2s=(a+b+c)/2 and

θ(a,b,c)=(−1)s[s+1]![s−a]![s−b]![s−c]![a]![b]![c]!;\theta(a,b,c)=(-1)^s\frac{[s+1]![s-a]![s-b]![s-c]!}{[a]![b]![c]!};θ(a,b,c)=(−1)s[a]![b]![c]![s+1]![s−a]![s−b]![s−c]!​;

set θ=0\theta=0θ=0 otherwise.

Whiskers A◃fA\triangleleft fA◃f and f▹Af\triangleright Af▹A mean id⁡A⊗f\operatorname{id}_A\otimes fidA​⊗f and f⊗id⁡Af\otimes\operatorname{id}_Af⊗idA​. With A=Va,B=Vb,C′=Vc,L=VlA=V_a,B=V_b,C'=V_c,L=V_lA=Va​,B=Vb​,C′=Vc​,L=Vl​, the upward vertex, split, and fuse are the following ordered composites, with only canonical regrouping suppressed:

vabc=uc≫(C′◃(ub▹C′))≫(C′◃((B◃(ua▹B))▹C′))≫(vcba▹(A⊗B⊗C′)),split⁡l;a,b=(ua▹L)≫(A◃((ub▹A)▹L))≫(A◃B◃vbal),fuse⁡a,b;l=((A⊗B)◃ul)≫(vabl▹L).\begin{aligned} v^{abc}&=u_c\gg(C'\triangleleft(u_b\triangleright C'))\gg \bigl(C'\triangleleft((B\triangleleft(u_a\triangleright B))\triangleright C')\bigr) \gg(v_{cba}\triangleright(A\otimes B\otimes C')),\\ \operatorname{split}_{l;a,b}&=(u_a\triangleright L)\gg \bigl(A\triangleleft((u_b\triangleright A)\triangleright L)\bigr) \gg(A\triangleleft B\triangleleft v_{bal}),\\ \operatorname{fuse}_{a,b;l}&=((A\otimes B)\triangleleft u_l)\gg(v_{abl}\triangleright L). \end{aligned}vabcsplitl;a,b​fusea,b;l​​=uc​≫(C′◃(ub​▹C′))≫(C′◃((B◃(ua​▹B))▹C′))≫(vcba​▹(A⊗B⊗C′)),=(ua​▹L)≫(A◃((ub​▹A)▹L))≫(A◃B◃vbal​),=((A⊗B)◃ul​)≫(vabl​▹L).​

Vertex normalization assumes exactly vabc≫vabc=ι(θ(a,b,c))id⁡1v^{abc}\gg v_{abc}=\iota(\theta(a,b,c))\operatorname{id}_{\mathbf1}vabc≫vabc​=ι(θ(a,b,c))id1​, and that

(ua▹(B⊗C′⊗A))≫(A◃(vabc▹A))≫ea=vbca(u_a\triangleright(B\otimes C'\otimes A))\gg \bigl(A\triangleleft(v_{abc}\triangleright A)\bigr)\gg e_a=v_{bca}(ua​▹(B⊗C′⊗A))≫(A◃(vabc​▹A))≫ea​=vbca​

for every natural triple. There is no tetrahedral seed hypothesis, no all-channel identification with a scalar tetrahedral sum, and no assumed general tetrahedral symmetry. No braiding or Hermitian structure is assumed.

Fix arbitrary a,b,c,d∈Na,b,c,d\in\mathbb Na,b,c,d∈N, with no admissibility restriction, and write Babcd=Va⊗Vb⊗Vc⊗VdB_{abcd}=V_a\otimes V_b\otimes V_c\otimes V_dBabcd​=Va​⊗Vb​⊗Vc​⊗Vd​, left-associated. The states and covectors are

Xl=ul≫(split⁡l;a,b⊗split⁡l;c,d),XlR=(fuse⁡a,b;l⊗fuse⁡c,d;l)≫el,Hj=vajd≫(Va◃(split⁡j;b,c▹Vd)),HjR=(Va◃(fuse⁡b,c;j▹Vd))≫vajd.\begin{aligned} X_l&=u_l\gg(\operatorname{split}_{l;a,b}\otimes\operatorname{split}_{l;c,d}),& X_l^R&=(\operatorname{fuse}_{a,b;l}\otimes\operatorname{fuse}_{c,d;l})\gg e_l,\\ H_j&=v^{ajd}\gg\bigl(V_a\triangleleft(\operatorname{split}_{j;b,c}\triangleright V_d)\bigr),& H_j^R&=\bigl(V_a\triangleleft(\operatorname{fuse}_{b,c;j}\triangleright V_d)\bigr)\gg v_{ajd}. \end{aligned}Xl​Hj​​=ul​≫(splitl;a,b​⊗splitl;c,d​),=vajd≫(Va​◃(splitj;b,c​▹Vd​)),​XlR​HjR​​=(fusea,b;l​⊗fusec,d;l​)≫el​,=(Va​◃(fuseb,c;j​▹Vd​))≫vajd​.​

States go from 1\mathbf11 to BabcdB_{abcd}Babcd​, covectors in the reverse direction. The superscript RRR denotes these explicitly defined covectors, not a conjugate-linear adjoint.

Define rational-function norms

hj={θ(a,d,j)θ(b,c,j)/Δ(j),(a,d,j),(b,c,j) admissible,0,otherwise,xl={θ(a,b,l)θ(c,d,l)/Δ(l),(a,b,l),(c,d,l) admissible,0,otherwise.h_j=\begin{cases}\theta(a,d,j)\theta(b,c,j)/\Delta(j),& (a,d,j),(b,c,j)\text{ admissible},\\0,&\text{otherwise},\end{cases} \quad x_l=\begin{cases}\theta(a,b,l)\theta(c,d,l)/\Delta(l),& (a,b,l),(c,d,l)\text{ admissible},\\0,&\text{otherwise}.\end{cases}hj​={θ(a,d,j)θ(b,c,j)/Δ(j),0,​(a,d,j),(b,c,j) admissible,otherwise,​xl​={θ(a,b,l)θ(c,d,l)/Δ(l),0,​(a,b,l),(c,d,l) admissible,otherwise.​

They are bilinear pairing scalars, not positive or Hermitian norms.

The eight conclusions

For every fixed natural quadruple (a,b,c,d)(a,b,c,d)(a,b,c,d), the assertion is the conjunction of the following eight statements.

  1. For every j,k∈Nj,k\in\mathbb Nj,k∈N,
Hk≫HjR={ι(hj)id⁡1,k=j,0,k≠j.H_k\gg H_j^R=\begin{cases}\iota(h_j)\operatorname{id}_{\mathbf1},&k=j,\\0,&k\ne j.\end{cases}Hk​≫HjR​={ι(hj​)id1​,0,​k=j,k=j.​
  1. For every l,m∈Nl,m\in\mathbb Nl,m∈N,
Xm≫XlR={ι(xl)id⁡1,m=l,0,m≠l.X_m\gg X_l^R=\begin{cases}\iota(x_l)\operatorname{id}_{\mathbf1},&m=l,\\0,&m\ne l.\end{cases}Xm​≫XlR​={ι(xl​)id1​,0,​m=l,m=l.​
  1. For every state f:1→Babcdf:\mathbf1\to B_{abcd}f:1→Babcd​, there exists a function s:N→Ks:\mathbb N\to Ks:N→K such that, simultaneously,
∀j∈N,f≫HjR=s(j)id⁡1,f=∑j=0b+cs(j)ι(hj)Hj.\forall j\in\mathbb N,\quad f\gg H_j^R=s(j)\operatorname{id}_{\mathbf1},\qquad f=\sum_{j=0}^{b+c}\frac{s(j)}{\iota(h_j)}H_j.∀j∈N,f≫HjR​=s(j)id1​,f=j=0∑b+c​ι(hj​)s(j)​Hj​.
  1. For every covector g:Babcd→1g:B_{abcd}\to\mathbf1g:Babcd​→1, there exists a function s:N→Ks:\mathbb N\to Ks:N→K such that
∀j∈N,Hj≫g=s(j)id⁡1,g=∑j=0b+cs(j)ι(hj)HjR.\forall j\in\mathbb N,\quad H_j\gg g=s(j)\operatorname{id}_{\mathbf1},\qquad g=\sum_{j=0}^{b+c}\frac{s(j)}{\iota(h_j)}H_j^R.∀j∈N,Hj​≫g=s(j)id1​,g=j=0∑b+c​ι(hj​)s(j)​HjR​.
  1. For every state f:1→Babcdf:\mathbf1\to B_{abcd}f:1→Babcd​, there exists a function s:N→Ks:\mathbb N\to Ks:N→K such that
∀l∈N,f≫XlR=s(l)id⁡1,f=∑l=0a+bs(l)ι(xl)Xl.\forall l\in\mathbb N,\quad f\gg X_l^R=s(l)\operatorname{id}_{\mathbf1},\qquad f=\sum_{l=0}^{a+b}\frac{s(l)}{\iota(x_l)}X_l.∀l∈N,f≫XlR​=s(l)id1​,f=l=0∑a+b​ι(xl​)s(l)​Xl​.
  1. For every covector g:Babcd→1g:B_{abcd}\to\mathbf1g:Babcd​→1, there exists a function s:N→Ks:\mathbb N\to Ks:N→K such that
∀l∈N,Xl≫g=s(l)id⁡1,g=∑l=0a+bs(l)ι(xl)XlR.\forall l\in\mathbb N,\quad X_l\gg g=s(l)\operatorname{id}_{\mathbf1},\qquad g=\sum_{l=0}^{a+b}\frac{s(l)}{\iota(x_l)}X_l^R.∀l∈N,Xl​≫g=s(l)id1​,g=l=0∑a+b​ι(xl​)s(l)​XlR​.

In statements 3–6 the existential functions are independently chosen for the given morphism and expansion. The pairing equalities cover all natural indices, even outside the finite sum. The sums range over all integers from zero through the displayed inclusive upper endpoint, not filtered admissible ranges. The statement does not explicitly assert uniqueness of these functions or nonzero norms at every summed index. 7. For every j,l∈Nj,l\in\mathbb Nj,l∈N,

Xl(a,b,c,d)≫HjR(a,b,c,d)=Tet⁡(a,b,c,d;j,l).X_l(a,b,c,d)\gg H_j^R(a,b,c,d)=\operatorname{Tet}(a,b,c,d;j,l).Xl​(a,b,c,d)≫HjR​(a,b,c,d)=Tet(a,b,c,d;j,l).

The tetrahedral endomorphism is defined, for integer channels r,tr,tr,t, by

Tet⁡(a,b,c,d;r,t)={Xr(a,d,c,b)≫HtR(a,d,c,b),r,t≥0,0,otherwise.\operatorname{Tet}(a,b,c,d;r,t)= \begin{cases} X_r(a,d,c,b)\gg H_t^R(a,d,c,b),&r,t\ge0,\\0,&\text{otherwise}. \end{cases}Tet(a,b,c,d;r,t)={Xr​(a,d,c,b)≫HtR​(a,d,c,b),0,​r,t≥0,otherwise.​

Thus the literal unfolded assertion here, since j,lj,lj,l are natural, is

Xl(a,b,c,d)≫HjR(a,b,c,d)=Xj(a,d,c,b)≫HlR(a,d,c,b).\boxed{X_l(a,b,c,d)\gg H_j^R(a,b,c,d) =X_j(a,d,c,b)\gg H_l^R(a,d,c,b).}Xl​(a,b,c,d)≫HjR​(a,b,c,d)=Xj​(a,d,c,b)≫HlR​(a,d,c,b).​

More explicitly, the left composite is

ul≫(split⁡l;a,b⊗split⁡l;c,d)≫(Va◃(fuse⁡b,c;j▹Vd))≫vajd,u_l\gg(\operatorname{split}_{l;a,b}\otimes\operatorname{split}_{l;c,d}) \gg\bigl(V_a\triangleleft(\operatorname{fuse}_{b,c;j}\triangleright V_d)\bigr)\gg v_{ajd},ul​≫(splitl;a,b​⊗splitl;c,d​)≫(Va​◃(fuseb,c;j​▹Vd​))≫vajd​,

whereas the right composite is

uj≫(split⁡j;a,d⊗split⁡j;c,b)≫(Va◃(fuse⁡d,c;l▹Vb))≫valb.u_j\gg(\operatorname{split}_{j;a,d}\otimes\operatorname{split}_{j;c,b}) \gg\bigl(V_a\triangleleft(\operatorname{fuse}_{d,c;l}\triangleright V_b)\bigr)\gg v_{alb}.uj​≫(splitj;a,d​⊗splitj;c,b​)≫(Va​◃(fused,c;l​▹Vb​))≫valb​.

This is a substantive equality between closures with boundary orders (a,b,c,d)(a,b,c,d)(a,b,c,d) and (a,d,c,b)(a,d,c,b)(a,d,c,b) and interchanged channel roles, not merely the defining formula for Tet. Pure substitution into that definition instead gives the left closure as Tet⁡(a,d,c,b;l,j)\operatorname{Tet}(a,d,c,b;l,j)Tet(a,d,c,b;l,j); the assertion therefore equates that endomorphism with Tet⁡(a,b,c,d;j,l)\operatorname{Tet}(a,b,c,d;j,l)Tet(a,b,c,d;j,l). 8. For every l∈Nl\in\mathbb Nl∈N, there exists a function s:N→Ks:\mathbb N\to Ks:N→K such that

∀j∈N,Tet⁡(a,b,c,d;j,l)=s(j)id⁡1,Xl(a,b,c,d)=∑j=0b+cs(j)ι(hj)Hj(a,b,c,d).\forall j\in\mathbb N,\quad \operatorname{Tet}(a,b,c,d;j,l)=s(j)\operatorname{id}_{\mathbf1},\qquad X_l(a,b,c,d)=\sum_{j=0}^{b+c}\frac{s(j)}{\iota(h_j)}H_j(a,b,c,d).∀j∈N,Tet(a,b,c,d;j,l)=s(j)id1​,Xl​(a,b,c,d)=j=0∑b+c​ι(hj​)s(j)​Hj​(a,b,c,d).

This expands each vertical state in horizontal states, not the reverse expansion. It asserts scalar existence for every Tet in this natural-channel family, but does not give a numerical formula for those scalars or identify them with a separately defined finite-sum tetrahedral evaluation. The same function supplies both all-channel scalar equalities and the finite expansion, and may depend on a,b,c,d,la,b,c,d,la,b,c,d,l.

All divisions are total divisions in a rational-function field, so a zero denominator yields zero. No nonvanishing hypotheses or restrictions to a complex specialization are added. These are pairings and spanning assertions for the full natural-indexed families, not a claim that every indexed state is a nonzero basis element. The tetrahedral symmetry in statement 7 is among the conclusions to establish, not among the assumptions.

Preamble
import Definitions.Def_QuantumSixJNetwork

open CategoryTheory MonoidalCategory
open scoped BigOperators QuantumSixJ

universe v u
Formal statement
namespace QuantumSixJ.Network

theorem four_valent_bases {C : Type u} [Category.{v} C] [Preadditive C] [Linear K C]
    [MonoidalCategory C] [MonoidalPreadditive C] [MonoidalLinear K C]
    (S : SuQ2Data C) (W : Vertex S) (hS : SuQ2Axioms S) (hW : VertexNormalized W) :
    ∀ a b c d : ℕ,
      (∀ j k : ℕ, W.H a b c d k ≫ W.Hcov a b c d j =
        if k = j then ι (h a b c d j) • 𝟙 (𝟙_ C) else 0) ∧
      (∀ l m : ℕ, W.X a b c d m ≫ W.Xcov a b c d l =
        if m = l then ι (x a b c d l) • 𝟙 (𝟙_ C) else 0) ∧
      (∀ f : 𝟙_ C ⟶ S.V a ⊗ S.V b ⊗ S.V c ⊗ S.V d,
        ∃ s : ℕ → K,
          (∀ j : ℕ, f ≫ W.Hcov a b c d j = s j • 𝟙 (𝟙_ C)) ∧
          f = ∑ j ∈ Finset.range (b + c + 1), (s j / ι (h a b c d j)) • W.H a b c d j) ∧
      (∀ g : S.V a ⊗ S.V b ⊗ S.V c ⊗ S.V d ⟶ 𝟙_ C,
        ∃ s : ℕ → K,
          (∀ j : ℕ, W.H a b c d j ≫ g = s j • 𝟙 (𝟙_ C)) ∧
          g = ∑ j ∈ Finset.range (b + c + 1), (s j / ι (h a b c d j)) • W.Hcov a b c d j) ∧
      (∀ f : 𝟙_ C ⟶ S.V a ⊗ S.V b ⊗ S.V c ⊗ S.V d,
        ∃ s : ℕ → K,
          (∀ l : ℕ, f ≫ W.Xcov a b c d l = s l • 𝟙 (𝟙_ C)) ∧
          f = ∑ l ∈ Finset.range (a + b + 1), (s l / ι (x a b c d l)) • W.X a b c d l) ∧
      (∀ g : S.V a ⊗ S.V b ⊗ S.V c ⊗ S.V d ⟶ 𝟙_ C,
        ∃ s : ℕ → K,
          (∀ l : ℕ, W.X a b c d l ≫ g = s l • 𝟙 (𝟙_ C)) ∧
          g = ∑ l ∈ Finset.range (a + b + 1), (s l / ι (x a b c d l)) • W.Xcov a b c d l) ∧
      (∀ j l : ℕ, W.X a b c d l ≫ W.Hcov a b c d j = W.tet a b c d j l) ∧
      (∀ l : ℕ, ∃ s : ℕ → K,
        (∀ j : ℕ, W.tet a b c d j l = s j • 𝟙 (𝟙_ C)) ∧
        W.X a b c d l = ∑ j ∈ Finset.range (b + c + 1),
          (s j / ι (h a b c d j)) • W.H a b c d j) := by sorry

end QuantumSixJ.Network
Source
I. Khavkine, Recurrence relation for the 6j-symbol of su_q(2) as a symmetric eigenvalue problem, International Journal of Geometric Methods in Modern Physics (2015), https://doi.org/10.1142/S0219887815501170; https://arxiv.org/abs/1009.2261. Eqs. (5)–(10), coupled four-valent networks, graphical Tet, orthogonality, norms and recoupling expansion.
Human review
  • Endorsed by Shuze Chen · Nov 19, 2026

    Confirmed by the moderator at approval.

  • Endorsed by igor.kh · Nov 19, 2026

    Confirmed by the mission captain (proposal self-audit).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me