Resolution criteria

This market resolves to YES if a fatal mathematical flaw is identified and verified in the preprint "The Width of the Conformal Fan: Dependence and the Variance of Realized Coverage" (https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6979679) by Peter Cotton by EOY 2026.

A "fatal mathematical flaw" is defined as an error that invalidates the paper's primary novel claims, such as the counterexample presented in Section 3 or the proof of Theorem 1.

Evidence of a fatal error includes:

1. The preprint is officially retracted or marked as "withdrawn" on arXiv.

2. The author publishes an updated version or erratum on arXiv explicitly acknowledging a fatal flaw in the core proofs, counterexamples, or theorems.

3. An academic publication or preprint by established researchers in the conformal prediction field provides a mathematically verified counterproof or counterexample demonstrating that Sornwanee's primary results are incorrect, without a successful rebuttal.

This market resolves to NO if no such fatal flaw is verified, or if the paper is peer-reviewed and accepted at a major ML/statistics venue (such as NeurIPS, ICML, or AISTATS) without its core mathematical contributions being discredited.

The market creator should adjust the target cutoff date (e.g., December 31, 2026) as needed to allow sufficient time for academic review, and will make the final determination based on academic consensus (e.g., discussions on OpenReview, arXiv updates, or peer-reviewed comments).