Li (2026), ‘arXiv preprint, cs.LG. v1 (2026-03-16) is the version read here. v2 (2026-09-14) under the same identifier is a different paper, “The Metric Slingshot: Navigational Reuse as Width-Optimal Structural Decoupling in Continual Learning”, also read; see corrections.’ — arXiv:2603.15412
Standing in the record
Filed on 2026-10-01 at the owner’s request.
published: is the first appearance (ADR-002). This entry is arXiv v1 (2026-03-16), “Local Urysohn Width”, the paper the owner asked for. Since 2026-09-14 the identifier serves v2, “The Metric Slingshot”, a different paper on grid cells and navigation with no shared theorems; the reading (NOTE) covers v2 in its own section, where it would be filed Rejected. Read together with LIT-373 and LIT-374 as one research programme (curation entry of 2026-10-01).
It was filed Deferred, unread. NOTE-323 is the close reading of 2026-10-01, and it placed the work: Proposed — >- Promising and unproven. It is the cleanest and most honest paper in Li’s programme: definitions are precise, the main theorems are correctly proved, a dependency table separates proved from conjectural results, and earlier errors are corrected openly. It is also the consistent construction that LIT-374’s and arXiv 2602.07974’s bouquet lacked. What would settle it: a result showing that this width governs error or sample cost for learners not built from local experts (for example local-learning or nearest-neighbour rates), or a proof of Conj 8.9 under conditions its motivating examples actually satisfy. Without one of these, it is a covering number with a topological headline.