ModelingPaperKit

Design simple, defensible baseline models for CUMCM prediction, classification, evaluation, optimization, simulation, and ranking tasks before using stronger methods. Use when a team needs a minimum viable model, method comparison, ablation baseline, or fallback path under contest time pressure.

Enforce contest-time safety for CUMCM work, including no public problem discussion browsing, no outside solution sharing, source citation discipline, AI-use logging, instructor-contact boundaries, and submission-risk warnings. Use during the active CUMCM contest window or when handling current contest problem statements.

Audit CUMCM datasets for missing values, units, duplicates, outliers, inconsistent identifiers, impossible ranges, time/order problems, leakage, and preprocessing decisions. Use when reviewing raw or processed data before modeling, writing data cleaning notes, or deciding whether results are trustworthy.

Maintain CUMCM experiment records for datasets, preprocessing versions, model parameters, random seeds, metrics, generated figures, result tables, and paper claims. Use when tracking runs, comparing model variants, reproducing outputs, or preparing appendix and supporting-material logs.

Parse and decompose 2026 CUMCM problem statements into subquestions, data requirements, deliverables, hidden constraints, scoring signals, assumptions to verify, and first-day action plans. Use when reading a new CUMCM problem, extracting tasks, clarifying what must be modeled, or building the initial problem brief before choosing methods.

Check consistency between CUMCM paper claims, formulas, tables, figures, units, code outputs, appendix records, and subproblem conclusions. Use when reviewing a draft for mismatched numbers, stale figures, inconsistent symbols, unsupported conclusions, or broken result narrative.

Audit CUMCM supporting materials for required code, data, generated outputs, AI-use details, appendix file lists, package size, cache files, identity leaks, and paper-to-file traceability. Use when preparing ZIP/RAR support packages or checking whether materials can reproduce paper results.

Design and review validation, robustness, sensitivity analysis, error analysis, and uncertainty checks for CUMCM models. Use when deciding how to prove model credibility, comparing parameter choices, checking stability, or writing validation and sensitivity sections.

Track, verify, and apply 2026 CUMCM / 高教社杯 official rules, notices, paper format requirements, AI-tool rules, schedule changes, and submission instructions. Use when asked to check latest contest rules, update repository guidance for 2026 CUMCM, compare templates with official requirements, or keep preparation work aligned with current CUMCM notices.

Run a final 2026 CUMCM readiness review across paper content, LaTeX build, template compliance, result consistency, figures/tables, citations, AI-use records, anonymity, supporting materials, and submission packaging. Use when the user asks for final review, pre-submit review, readiness check, or overall quality audit.

Prepare and review 2026 CUMCM electronic and paper submission packages, including final PDF, supporting materials, code/data folders, AI-use details, appendix file list, paper-only commitment and numbering pages, naming conventions, and last-mile safety checks. Use when packaging deliverables or making a submission checklist.

Structure, outline, and revise a 2026 CUMCM paper for strong mathematical-modeling narrative flow, including abstract, problem restatement, assumptions, notation, model construction, solution, results, validation, sensitivity, evaluation, conclusion, references, and appendix. Use when asked to plan sections, improve paper logic, rewrite section skeletons, or prepare a writing roadmap.

Design, review, and improve figures, tables, captions, result summaries, and visual evidence in 2026 CUMCM papers. Use when deciding what plots or tables to include, improving captions, checking figure/table references, converting results into competition-readable visuals, or aligning assets with ModelingPaperKit LaTeX macros.

Audit the ModelingPaperKit CUMCM LaTeX template for 2026 CUMCM paper-format compliance, electronic versus paper submission behavior, page order, margins, fonts, section structure, AI declaration placement, references, appendix, and build risks. Use when reviewing or editing templates/cumcm or preparing a template-quality pass.

Scan CUMCM papers, LaTeX files, metadata, code, paths, figures, data files, and supporting materials for identity leaks before 2026 CUMCM submission. Use when checking anonymity, removing school/team/member/advisor names, detecting phone/email/student IDs, or reviewing electronic submission safety.

Manage references, source records, citation discipline, and AI-tool usage logs for 2026 CUMCM papers and supporting materials. Use when checking citations, adding bibliographic entries, recording AI use, deciding where AI statements belong, or preparing source/AI compliance notes.

Organize reproducible data processing, modeling code, generated outputs, and supporting materials for 2026 CUMCM. Use when creating or reviewing code folders, data provenance, preprocessing scripts, experiment logs, generated figures/tables, appendix code listings, or reproducibility instructions.

Create rigorous modeling plans for 2026 CUMCM problems from problem statements, data descriptions, and constraints. Use when decomposing a CUMCM task, selecting candidate models, defining assumptions, metrics, algorithms, validation plans, sensitivity analysis, or contest-time modeling strategy.

Use this skill when working with the ModelingPaperKit repository to write, edit, compile, review, or prepare math modeling contest papers for CUMCM, MCM/ICM, Wuyi, Beijing, or the included example templates. Use it for LaTeX template edits, section organization, figure/table insertion, build-log diagnosis, final submission checks, contest safety constraints, AI usage/source logging, and PaperKit-specific workflows.