Complementary disclosures — v2.0.0 candidate

Complementary disclosures in financial networks

Version 2.0.0-candidate consolidates the supplied manuscript and its unpublished companion.

The revision corrects the zero-factor boundary in the prior-dependent submodularity criterion, incorporates the attributed one-layer construction, and distinguishes accounting coupling from nonlinear transmission. It adds affine-simplex identification and residual-width characterisations using standard LP duality, bounded-error affine certificates, and a full-range heterogeneous-error formula for the layered family.

The repaired computational study retains its complete 44-instance design, including unresolved cases. Rational witnesses are checked by exact clearing; retained UNSAT results remain solver evidence, not independently checked proof certificates. Historical results are preserved and labelled separately. See REPLAY.md, ASSURANCE.md, CLAIMS.md and ENVIRONMENT.json for exact counts and scope.

The package contains the scientific manuscript and PDF, code, retained computational records, semantic regression checks, AI index, provenance and component licences. GitHub and Zenodo assets are the same frozen files.

This is an unrefereed mathematical candidate about a specified synthetic model and report menu. No empirical banking validation, welfare conclusion, external peer-review acceptance, formal verification or official REF rating is claimed. Standard complementarity, clearing theory, loss conservation and LP duality are not claimed as new.

Anonymous. Original prose and data: CC0-1.0. Original code: MIT.

Version DOI: https://doi.org/10.5281/zenodo.23363516