The Setting
The configurations are field-valued labels on permutation words, with a distinguished cyclic sequence of labels. The constraints concern square gaps along this sequence and the dependence of one coordinate on all the others.
Fix , , and the block set . A permutation word on is a function assigning a permutation of to each block of .✦ The vertex set and the cyclic vertices are given by (1), where is the cyclic shift. (1) There are blocks.
Let be prime. Strict squares in are its nonzero squares; their set and the corresponding ordered pair space are defined in (2). (2) Every below is a uniform probability average, whereas denotes a counting sum. A quotient with denominator zero is assigned the value zero. For a real function on , the square pair average is the uniform average defined by (3). (3)✦
Component Averages
For , an inactive-block address on is a function assigning a permutation of to each block in .✦ The address space , list index set and list entry set are defined by (4); the projection restricts a vertex to its inactive blocks. (4) For a finite block set , a group , a shift and a nonnegative integer , threshold shift adjacency is the relation on group-valued addresses defined by (5). (5) Write for the list constraint edges. A field-valued list has square-constraint indicator as in (6); the nonzero square indicator is precisely when is a strict square and otherwise. (6) A list indexing selects the output configuration in (7). (7)✦
An ordered position pair is a pair with .✦ For a vertex , a block and positions , the position transposition is the word obtained by replacing with and leaving all other blocks fixed.✦ With , the transposition interaction is the matrix in (8). (8) For a real matrix on and indexings into a finite set , the label coincidence pairing , index energy and Boltzmann weight are defined in (9). The tuple weight uses and . (9) The tuple component average is the weighted average of a real observable defined by (10). (10)✦
The truncated family consists of block subsets of size at most ; the terminal family consists of those of size . For a family of block subsets, the restricted mixture and its normalized average use the block weights in (11). (11) The full mixture is , with normalizer .✦ The full tuple average and exceptional average use this normalizer, while the normalized good average uses its restricted normalizer, as in (12); denotes the exceptional mass. (12)
Cuts and Conditional Energy
For and distinct positions , the positive cut half consists of words in which precedes in block .✦ The reflection , for and distinct positions , transposes the values and in block , as specified by (13). (13) For and , the simultaneous square affine map on configurations is defined in (14). (14)
For a real linear functional on functions on a finite space and a bijection , the point mass and the joint density relative to uniform probability are defined in (15). (15) The coordinate decomposition , for , gives the splitting ; write . For a real function and a real kernel on , the conditional energy is the average over defined by (16). (16)✦
Fixed Thresholds
The pullback of a directed graph by a bijection is the directed graph whose edges are the pairs for which .✦ The crossing edges of a directed graph on are the pairs joined in either direction.✦ For a vertex of a finite directed graph , let be the number of crossing edges for the split into and its complement. The graph conditional energy constant is defined by (17). (17)✦ The leading list identification is the map , where is the unique index in . The leading constraint graph , its edge count , the leading mass scale and the mass threshold are specified in (18). (18) Put and .✦ The fixed Boltzmann bound , interaction scale , and marked and unmarked reflection constants and are given in (19). (19) The conditional error numerator and conditional decay constant are defined in (20). (20) The reflection threshold and conditional decay threshold are (21). (21)
The Claim
Theorem 1. Let be prime. If , the identities (22) and (23) hold: (22)✦ (23)✦ If , then for , distinct positions and every real , reflection positivity takes the form (24): (24)✦ For , and every real observable , the affine identities (25) and (26) hold, the latter for every : (25)✦ (26)✦ If , then every and real satisfy the selected-pair identity (27): (27)✦ Let have , let satisfy , and let . The output satisfies (28) for every : (28)✦ If , then every and real function of uniform mean zero satisfy the squared estimates (29) and (30): (29)✦ (30)✦ There is an integer such that every prime satisfies (31): (31)✦
In the paper, the result reads: there is an absolute integer such that every prime admits a probability law on and a nonnegative submeasure of mass , with , such that is reflection positive, both and are invariant under simultaneous for and , every one-label marginal of and is uniform, under all gaps for are strict squares and each ordered pair is uniform on , and for , every and every complex function of uniform mean zero, , where the left norm uses the other-label marginal and the right norm uses uniform probability on .
What Is Checked
The statements are over weighted permutation-word configurations in , with each assertion using its specified prime threshold. The component averages and the full mixture are linear in their real observable .