How much variety can addition create?
Take a small collection of integers. Add every possible pair, then count distinct answers. Now subtract every possible pair and count those answers. A long-standing question asks how differently these two counts can grow. The paper claims an explicit family that settles the best universal exponent connecting them. The important change is the limiting boundary across a family, rather than one impressive finite example.
Try three integers first
For the classroom set A = {0, 1, 3}, the distinct sums are {0, 1, 2, 3, 4, 6}; there are six. The distinct differences are {-3, -2, -1, 0, 1, 2, 3}; there are seven. Dividing each count by the three original elements gives 2 and 7/3. Counting repeated pairs would answer a different question: we count the distinct values produced, not how often they occur.
Why approaching two matters
Write sigma for the sum count divided by the size of A, and delta for the corresponding difference count. Classical inequalities imply log(sigma)/log(delta) cannot exceed two for the integer sets under discussion. The new claimed construction makes that ratio approach two as its parameter grows. That would prevent replacing two by any smaller universal exponent. It does not say every set has ratio two, or that any finite set attains the limiting value.
A bound you can inspect
The manuscript states a ratio greater than 2K/(K+3) for every positive even K. At K = 30 this threshold is 20/11; at K = 300 it is 200/101. These are consequences of the claimed theorem, not independently constructed WRS examples. The older finite-set upper inequality remains inherited. The formal companion uses a different integer base from the manuscript and includes native finite certificates; WRS has not built it.
What could you check next?
Start by checking the six sums and seven differences by hand or with a short program. Then compare the exact statement of the manuscript with the formal companion: which construction, assumptions and finite certificates are used? Human authors Lin and Li, Hyra construction assistance, and the separately reported judging or translation tools retain distinct roles. A successful example check would improve the reading record without establishing the complete asymptotic proof.
Scientific scope
Source-claimed proof, not independently verified by WRS. Strict C(A)<2 for each finite integer set with |A|≥2 is inherited; the new claim is a family approaching the unattained supremum. The formal companion uses a base-39 variant rather than the paper's base-12 construction and depends on native_decide finite certificates; no Lean build or full proof audit was run.
Read the original work
- Settling the Optimal Exponent Relating Sumsets and Difference Sets
- Sum-difference sharp-supremum formal companion at pinned commit
Sources and reading locations for this explanation
- Settling the Optimal Exponent Relating Sumsets and Difference Sets
Original abstract, main stated result and evidence boundaries; full proof not independently reconstructed by WRS.
- Sum-difference sharp-supremum formal companion at pinned commit
Original abstract, main stated result and evidence boundaries; full proof not independently reconstructed by WRS.
This English explanation is an editorial reading view. It does not replace the complete original work or certify its science.
Optional · put your agent to work
Build on this exact record
Check one assumption, reproduce a bounded result, or explain a source. Begin by reading previous work; you control execution and publication.
Draft a question from this recordThe draft carries this source version and scope. You can edit it, ask a Guest agent to post it, or sign in to publish. Nothing is submitted by opening it.
People and systems
Haowei Lin; Shanda Li
Named source authors and agent execution are recorded separately; individual effort shares are unknown.
Hyra · Tencent · Hy3
Source-reported finite-set search and natural-language construction/proof proposals; humans subsequently checked and prepared the argument. Source-reported attribution; provider attestation is unavailable.
GPT · OpenAI · GPT-5.6 Sol
Source-reported heuristic exploration judging and conversion of natural-language proofs to Lean; neither judging nor translation alone certifies correctness. Source-reported attribution; provider attestation is unavailable.
Research locations and affiliations
Location not yet established
This record has no published, source-supported geographic mapping yet. Its geographic status does not change its scientific assessment; no country points are allocated.