Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 5.1 (lattices) — fixed lattice with gap M1/s−εM^{1/s-\varepsilon}M1/s−ε in ℓp\ell_pℓp​ for any quadratic verifier over F2\mathbb{F}_2F2​, explicit exponent

Proved
LatticePrepro.lattice_gap

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

coding-theorycvp-ncp-preprocessinghardness-of-approximationlatticespreprocessing

Let 1≤p≤∞1\le p\le\infty1≤p≤∞ be a norm exponent (any real p≥1p\ge1p≥1, or p=∞p=\inftyp=∞), and let s=min⁡{p,2}s=\min\{p,2\}s=min{p,2}, with s=2s=2s=2 for p=∞p=\inftyp=∞. Let 0<ε<1/s0<\varepsilon<1/s0<ε<1/s. Let VVV be a quadratic verifier over F2\mathbb{F}_2F2​ with instance length nnn, SSS assignment coordinates and language L(V)⊆{0,1}nL(V)\subseteq\{0,1\}^nL(V)⊆{0,1}n, and let

C=sizeExp(2, 1−sε),C=\mathrm{sizeExp}(2,\,1-s\varepsilon),C=sizeExp(2,1−sε),

an explicit exponent depending only on sεs\varepsilonsε. Then there are a dimension MMM, a lattice Λ⊆ZM\Lambda\subseteq\mathbb{Z}^MΛ⊆ZM (a Z\mathbb{Z}Z-submodule), an integer target tx∈ZMt_x\in\mathbb{Z}^Mtx​∈ZM for every x∈{0,1}nx\in\{0,1\}^nx∈{0,1}n, an integer radius R≥1R\ge1R≥1 and an integer N≥1N\ge1N≥1 such that:

  1. 1≤M≤(S+1)C1\le M\le(S+1)^C1≤M≤(S+1)C;
  2. N⋅ZM⊆ΛN\cdot\mathbb{Z}^M\subseteq\LambdaN⋅ZM⊆Λ, so Λ\LambdaΛ has full rank;
  3. for every x∈L(V)x\in L(V)x∈L(V), dist⁡p(tx,Λ)≤R\operatorname{dist}_p(t_x,\Lambda)\le Rdistp​(tx​,Λ)≤R;
  4. for every x∉L(V)x\notin L(V)x∈/L(V), dist⁡p(tx,Λ)>M1/s−ε⋅R\operatorname{dist}_p(t_x,\Lambda)>M^{1/s-\varepsilon}\cdot Rdistp​(tx​,Λ)>M1/s−ε⋅R.

Here ∥⋅∥p\|\cdot\|_p∥⋅∥p​ is the ℓp\ell_pℓp​ norm on RM\mathbb{R}^MRM (the maximum norm for p=∞p=\inftyp=∞) and dist⁡p(t,Λ)=inf⁡v∈Λ∥t−v∥p\operatorname{dist}_p(t,\Lambda)=\inf_{v\in\Lambda}\|t-v\|_pdistp​(t,Λ)=infv∈Λ​∥t−v∥p​. In short,

x∈L(V)⇒dist⁡p(tx,Λ)≤R,x∉L(V)⇒dist⁡p(tx,Λ)>M1/s−ε⋅R.x\in L(V)\Rightarrow\operatorname{dist}_p(t_x,\Lambda)\le R,\qquad x\notin L(V)\Rightarrow\operatorname{dist}_p(t_x,\Lambda)>M^{1/s-\varepsilon}\cdot R.x∈L(V)⇒distp​(tx​,Λ)≤R,x∈/L(V)⇒distp​(tx​,Λ)>M1/s−ε⋅R.

The lattice Λ\LambdaΛ and the radius RRR do not depend on xxx; only the target txt_xtx​ does.

This is the lattice case of Theorem 5.1. As in the paper's proof, the verifier is over F2\mathbb{F}_2F2​. The result is stated for an arbitrary quadratic verifier, with dimension bounded by (S+1)C(S+1)^C(S+1)C. The paper fixes a 3SAT verifier and obtains M=poly(n)M=\mathrm{poly}(n)M=poly(n); that instantiation is a separate Lean result. The paper allows rational ppp; here every real p≥1p\ge1p≥1 is allowed.

Formalization Note The conclusion is Nonempty (LatticeGap pn ε V.Language (V.S + 1) C), where pn : PNorm encodes ppp. The lattice distance is a real infimum (⨅ over lattice vectors), and M1/s−εM^{1/s-\varepsilon}M1/s−ε is a real power (Real.rpow). Polynomial-time computability and description lengths are not formalized; the search test of Lemma 5.2 is stated separately (LatticeReduction.accepts_of_isApprox, LatticeReduction.not_accepts_of_isNo).

Note (existence form). This statement asserts only that some lattice with targets has the gap. The Lean statement does not require the targets to be computable, so this form is also satisfied by a construction unrelated to the paper, in which the target records whether the instance is in the language. The statement for the object that the paper constructs is LatticePrepro.QuadraticVerifier.Tables.latticeFamily_hasGap.

Preamble
import Mathlib
import Definitions.Def_LatticePrepro_Basic_Blocks
import Definitions.Def_LatticePrepro_Reduction_LinearSystem
import Definitions.Def_LatticePrepro_Conversions_Codes
import Definitions.Def_LatticePrepro_Conversions_ConstructionA
import Definitions.Def_LatticePrepro_Conversions_Hadamard
import Definitions.Def_LatticePrepro_Conversions_Lattices
import Definitions.Def_LatticePrepro_Lattices_Distance
import Definitions.Def_LatticePrepro_Lattices_Mathlib
import Definitions.Def_LatticePrepro_Moments_Basic
import Definitions.Def_LatticePrepro_Moments_Representation
import Definitions.Def_LatticePrepro_Verifier_Quadratic
import Definitions.Def_LatticePrepro_Reduction_FixedSystem
import Definitions.Def_LatticePrepro_Tables_Digits
import Definitions.Def_LatticePrepro_Tables_FieldSetup
import Definitions.Def_LatticePrepro_Tables_Hankel
import Definitions.Def_LatticePrepro_Tables_Construction
import Definitions.Def_LatticePrepro_Tables_ElementTable
import Definitions.Def_LatticePrepro_Tables_TableSoundness
import Definitions.Def_LatticePrepro_Tables_Explicit
import Definitions.Def_LatticePrepro_Main

open LatticePrepro
Formal statement
theorem LatticePrepro.lattice_gap (pn : PNorm) (ε : ℝ) (hε : 0 < ε) (hε1 : ε < 1 / pn.s)
    (V : QuadraticVerifier 2) :
    Nonempty (LatticeGap pn ε V.Language (V.S + 1) (sizeExp 2 (1 - pn.s * ε))) := by sorry
Source
G. Fenzi, G. Weissenberg, Nearly Optimal Hardness for CVP and NCP with Preprocessing (STOC 2027 submission #847; IACR ePrint 2026/2634, https://eprint.iacr.org/2026/2634), Theorem 5.1 (\label{thm:approximation-hardness}), lattice case; Lean: LatticePrepro/Main.lean, LatticePrepro.lattice_gap
Human review
  • Endorsed by Shuze Chen · Nov 15, 2026

    Confirmed by the moderator at approval.

  • Endorsed by WizardOfMenlo · Nov 15, 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