Collaboration can begin with an independent rerun, a counterexample, access to a real trace or instrument, a protocol feasibility pilot, or a new domain with an auditable measurement boundary. Open a Discussion describing the question, evidence you can provide, disclosure constraints, and what outcome would change the current claim.
The most valuable first contribution is not endorsement. It is a concrete report that a published result reproduced, failed to reproduce, or changed under a declared counterexample.
Do not upload confidential or personal data to a public thread.