Answers are grounded in this paper and cite their sources.
Preprint examines total-variation projection in temporal interfaces, suggesting implications for causal interpretation.
This preprint studies the identification problem for a complete finite temporal interface comprising an earlier event, a later intervention, a complete declared earlier transcript, and all remaining declared outcomes, with chronology and transcript immutability supplied by a separate physical passport. It compares the complete all-trial intervention law with a deliberately broad forward-completion class allowing arbitrary finite hidden memory, common causes, stochasticity, and command-dependent later processing, constrained only by independence of the mechanism producing the earlier transcript from the later command. The principal result is an exact, attained max-command total-variation projection onto this broad forward class. The complete-table distance reduces to the minimax total-variation radius of the command-indexed earlier marginals, and every minimizing center lifts to an optimal forward-completable table that preserves the observed later conditional distributions on positive rows. Supporting results characterize the response-kernel completion fiber through finite barycentric measures and use a spectator-extended OCB construction to illustrate the separation between no-backwards-in-time-signaling compatibility and definite causal order. The paper also delineates the independent evidence required before table-level incompatibility can support a physical backward-signaling interpretation, including verified chronology, immutable record commitment, intervention exogeneity, all-trial retention, ancillary and memory closure, ordinary-route closure, and finite-sample robustness. Status: Preprint; not peer reviewed.
No takes yet. Share an insight, caveat, or question.
Undisclosed Researcher (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: