Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Collections

ACM STOC 2027

Formalizations of theorems from papers submitted to ACM STOC 2027 (the 59th Annual ACM Symposium on Theory of Computing). Paper list: https://sigact-stoc.github.io/stoc2027/ . Missions here state a main result of a STOC 2027 paper in Lean 4 with Mathlib; each mission's title begins with the paper title.

178 completed missions

Missions

101–125 of 178
Open1321Completed178All1499
🏆Completed
CombinatoricsGraph TheoryLinear algebra·Captain: iris

On Brouwer's Laplacian conjectureResearch Paper

The Laplacian L=D−AL=D-AL=D−A of a graph records its degrees on the diagonal and its edges off it. Its spectrum controls connectivity, expansion, random walks, cuts and spanning trees, and sums of its largest eigenvalues appear in graph partitioning and in the energy of quadratic forms over a graph. A natural question is how large the top of the spectrum can be in…

19 thms3 active usersReviewed
🏆Completed
CombinatoricsGraph TheoryTheoretical Computer Science·Captain: iris

Tangle structure treesResearch Paper

Tangles were introduced by Robertson and Seymour in the Graph Minors project as a way to describe highly connected regions of a graph indirectly: instead of naming a vertex set, a tangle orients every low-order separation of the graph towards the side where the region lies. Diestel and coauthors later abstracted the notion away from graphs. In an abstract…

37 thms3 active usersReviewed
🏆Completed
CombinatoricsMachine LearningTheoretical Computer Science·Captain: iris

On the VC Dimension of Autoregressive Linear Threshold FunctionsResearch Paper

An autoregressive model applies one fixed next-token rule to the current sequence, appends the produced token, and repeats. Intermediate tokens act as computation steps, which is one explanation of why chain-of-thought generation increases the power of language models. Joshi et al. introduced a learning framework for this process: a fixed generator runs for …

22 thms3 active usersReviewed
🏆Completed
Complexity TheoryTheoretical Computer Science·Captain: iris

On the Composition Theorem for Approximate Degree, Randomized and Exact Quantum Query ComplexityResearch Paper

The approximate degree of a Boolean function f:{0,1}n→{0,1}f:\{0,1\}^n\to\{0,1\}f:{0,1}n→{0,1} is the least degree of a real polynomial that agrees with fff to within 1/31/31/3 at every point of the Boolean cube. It lower-bounds bounded-error quantum query complexity, and through the pattern-matrix and related methods it drives lower bounds in communication complexity, circuit complexity and…

32 thms3 active usersReviewed
🏆Completed
AlgebraCombinatoricsComplexity Theory+1·Captain: iris

Exponential quantum speedup for F_3^n-Subset-Sum? Or, rigorous classical algorithms for Binary-Error LWEResearch Paper

The search for natural problems with an exponential quantum speedup has few candidates outside quantum simulation and hidden algebraic structure. Problems next to post-quantum cryptography are a natural place to look, because Regev's quantum reduction for Learning With Errors (Regev 2009) can be run "in reverse" to turn a classical decoder into a quantum…

36 thms3 active usersReviewed
🏆Completed
CombinatoricsGraph TheoryProbability+1·Captain: iris

Randomized Greedy MIS is Fast at Every NodeResearch Paper

A local computation algorithm answers a query about one part of a global solution (here: is vertex vvv in the maximal independent set?) by inspecting only a small neighbourhood of vvv, so that answers to different queries are consistent with one global solution. Nguyen and Onak (FOCS 2008) simulated the randomized greedy maximal independent set (RGMIS) this…

41 thms3 active usersReviewed
🏆Completed
CombinatoricsOperations ResearchOptimization+1·Captain: iris

Weighted Equitability and Matroid-Constrained DiscrepancyResearch Paper

Partitioning a set of goods, jobs or edges into structured pieces of nearly equal value is a recurring question in combinatorial optimization and fair division. Matroids are the standard abstraction of the structure involved: spanning trees of a graph, linearly independent column sets, and partition constraints are all bases of a matroid.

22 thms3 active usersReviewed
🏆Completed
Machine LearningStatistics·Captain: iris

Omniprediction for All Mixable Proper Losses and BeyondResearch Paper

A forecaster that outputs probabilities is usually trained against one loss function, yet its forecasts are often consumed by several downstream decision makers, each with its own loss, some of them unknown at training time. Omniprediction, introduced in the batch setting by Gopalan, Kalai, Reingold, Sharan and Wieder (arXiv:2109.05389) and in the online…

49 thms3 active usersReviewed
🏆Completed
CombinatoricsMachine Learning·Captain: mikedeng1

The Optimal Sample Complexity of Multiclass and List LearningResearch Paper

In multiclass classification, a prediction rule assigns one label to each data point, and the set of possible labels may be infinite. Learning theory asks how much labeled data suffices to predict accurately on new points. The number of labels alone does not describe the difficulty of a hypothesis class. A useful complexity parameter must also distinguish…

12 thms2 active usersReviewed
🏆Completed
AlgebraComputational ComplexityLinear algebra·Captain: mikedeng1

The sum-of-squares complexity of linear-algebraic principlesResearch Paper

A real matrix of rank greater than rrr cannot be a product of an n×rn\times rn×r matrix and an r×nr\times nr×n matrix. This elementary obstruction becomes a different question when the permitted certificate is a polynomial identity of bounded degree. Researchers studying algebraic proof complexity ask how much degree a specified certificate system needs to express such…

11 thms2 active usersReviewed
🏆Completed
Machine LearningProbabilityStatistics·Captain: Shuze Chen

The Higher-Order Identifiability Paradox: Learning and Certifying Hypergraph OrderResearch Paper

Predictors built on coarse views of a hypergraph — features that keep only interactions up to some order and discard the rest — are cheaper and more stable than full-order models, but discarding the higher-order part is safe only when that part carries little signal. Zhang, Yan, Qiao and Yu (The Higher-Order Identifiability Paradox: Learning and Certifying…

15 thms2 active usersReviewed
🏆Completed
CombinatoricsInformation TheoryProbability+1·Captain: Shuze Chen

Synchronization Strings over a Ternary AlphabetResearch Paper

Codes that correct substitutions and erasures assume that the receiver knows where each symbol belongs. Insertions and deletions break that assumption: a single deletion shifts every later symbol. Haeupler and Shahrasbi introduced synchronization strings (HS17) to restore position information cheaply. The sender attaches one symbol of a fixed string SSS to…

19 thms2 active usersReviewed
🏆Completed
Machine LearningProbabilityStatistics·Captain: mikedeng1

Calibrated Multiaccuracy: Deterministic and Randomized Sample ComplexityResearch Paper

Multiaccuracy asks that a predictor's residuals be uncorrelated with every test in a fixed class: subpopulations or features that an auditor can compute should not reveal systematic over- or under-prediction. The condition was introduced by Hébert-Johnson, Kim, Reingold and Rothblum (2018) under the name multi-AE, alongside multicalibration. Calibration asks…

25 thms2 active usersReviewed
🏆Completed
AlgebraNumerical AnalysisProbability+1·Captain: Shuze Chen

Approximate Syzygies of Smoothed Polynomials and ApplicationsResearch Paper

Many algorithms for tensor decomposition, polynomial system solving and learning mixtures reduce to linear algebra on Macaulay matrices: given polynomials b1,…,bmb_1,\dots,b_mb1​,…,bm​, the matrix of the map (h1,…,hm)↦∑ibihi(h_1,\dots,h_m)\mapsto\sum_i b_ih_i(h1​,…,hm​)↦∑i​bi​hi​ at a fixed output degree. Whether such an algorithm is robust to noise depends on how far this map is from losing rank, beyond the…

12 thms2 active usersReviewed
🏆Completed
Machine LearningTheoretical Computer Science·Captain: mikedeng1

Dense Language Generation Made Simple: Deterministic, Randomized, and Multi-Order AlgorithmsResearch Paper

Language generation in the limit asks whether a generator can eventually produce only new members of an unknown language after observing positive examples. This is a mathematical model of generation from examples; it imposes no computational resource bound. Eventual validity alone allows outputs to occupy a sparse part of the target. On the positive…

18 thms2 active usersReviewed
🏆Completed
Algorithmic Game TheoryComputational GeometryOperations Research·Captain: mikedeng1

Condorcet-Winning Sets and Peer Selection in Planar Metric ElectionsResearch Paper

A committee can satisfy a majority guarantee that no single candidate satisfies: no outside alternative is preferred to every committee member by more than half the electorate. Such a committee is a Condorcet-winning set, extending the single-candidate majority criterion studied in social choice. Elkind, Lang, and Saffidine introduced this set-valued…

21 thms2 active usersReviewed
🏆Completed
Computational ComplexityMachine Learning·Captain: mikedeng1

Depth Lower Bounds for ReLU Networks with Binary InputsResearch Paper

A ReLU network repeatedly forms affine combinations of available values and applies the function ReLU⁡(t)=max⁡(0,t)\operatorname{ReLU}(t)=\max(0,t)ReLU(t)=max(0,t). Its depth measures the number of computational layers, while its width measures the largest number of computed units in one layer. Researchers studying neural-network expressiveness ask how much width is needed when depth is…

10 thms2 active usersReviewed
🏆Completed
Discrete GeometryProbability·Captain: mikedeng1

Optimal Average Distortion for Non-Expansive Embeddings of Euclidean Metrics into the Line: With applications to cut-matching games and vertex expansionResearch Paper

Non-expansive maps of a metric space into the real line with small average distortion are a standard rounding device for graph partitioning. A relaxation (a linear or semidefinite program) produces a metric on the vertices; a 1-Lipschitz map to the line that preserves the average distance up to a factor DDD is then thresholded into a cut, and DDD becomes the…

9 thms2 active usersReviewed
🏆Completed
CombinatoricsMachine LearningModel Theory·Captain: mikedeng1

Encoding orders and trees in real-valued functionsResearch Paper

A bounded real-valued function can encode both branching choices and an ordered sequence of comparisons. Relating these two patterns matters in model theory, where the absence of long order patterns expresses stability, and in learning theory, where branching patterns measure sequential complexity. Conant and Terry study this relationship for functions…

16 thms2 active usersReviewed
🏆Completed
cryptographyNumber Theory·Captain: mikedeng1

Dual Coppersmith:Finding Small Roots Without Multivariate Polynomial SystemsResearch Paper

A bounded modular root is an integer solution of polynomial congruences whose coordinates lie in specified short intervals. Such problems arise in public-key cryptanalysis when an attacker knows algebraic relations involving a small unknown quantity. The usual multivariate Coppersmith workflow first uses lattice reduction to produce auxiliary integer…

10 thms2 active usersReviewed
🏆Completed
AlgebraComplexity TheoryLinear algebra+1·Captain: Shuze Chen

New lower bounds on tensor rank of (n,2,m)(n,2,m)(n,2,m) matrix multiplication with GPT-6 AstraResearch Paper

Every fast matrix multiplication algorithm starts from an identity that computes a fixed-shape product with few multiplications. Strassen's 1969 algorithm (Str69) multiplies two 2×22\times 22×2 matrices with 7 instead of 8 multiplications, and applying such an identity recursively turns RRR multiplications for a k×lk\times lk×l by l×ml\times ml×m product into an …

12 thms2 active usersReviewed
🏆Completed
CombinatoricsProbability·Captain: mikedeng1

Subgaussian Vector Balancing by an Ising ModelResearch Paper

Vector balancing asks how to choose signs for short vectors so that their sum remains controlled in every direction. A distribution over signs offers a probabilistic version: every linear measurement of the resulting signed sum should have a small exponential moment. This is useful when a randomized assignment must balance several numerical covariates…

17 thms2 active usersReviewed
🏆Completed
CombinatoricsMachine Learning·Captain: mikedeng1

Language Generation Beyond Adversarial ExamplesResearch Paper

Researchers in learning theory study whether examples of an unknown language suffice to generate new valid strings. The task asks for a new member of the target language after each received example, with errors allowed during an initial finite period. Language generation in the limit, introduced by Kleinberg and Mullainathan, supplies a mathematical model…

28 thms2 active usersReviewed
🏆Completed
AnalysisFunctional AnalysisProbability·Captain: mikedeng1

A dimension-free bound for the isoperimetric constant of unconditional log-concave measuresResearch Paper

A probability distribution in high dimension can be difficult to control even when every one-dimensional coordinate has variance one. A Poincaré inequality controls the variance of every locally Lipschitz real function by its average squared gradient. For log-concave distributions, this functional inequality also controls isoperimetry. It therefore connects…

24 thms2 active usersReviewed
🏆Completed
CombinatoricsTheoretical Computer Science·Captain: mikedeng1

A Tight Ω~(m)\widetilde\Omega (\sqrt m)Ω(m​) Information-theoretic Lower Bound for Randomized Online Set CoverResearch Paper

Online set cover asks how many sets must be purchased when the elements requiring coverage arrive one at a time. It is relevant to the analysis of algorithms that make irrevocable covering decisions before learning which demands will occur. Knowing the available sets does not reveal the future demands. The mathematical question is how much this missing…

17 thms2 active usersReviewed
PreviousPage 5 of 8Next

Top contributors

RankContributorAccepted solutionsSubmitted problems
1SCShuze Chen23927601
2DBdbenbenn909166
3SHshivm1630
4PAparclytaxel150172
5LRlreyzin140143
6KMkmehltretter8212219
7MTmartelus trismortuus1050
8BOBoomShaKaLaaaKa93100
9GAgaryholm900
10WCwcysai8486

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