{"id":"d710ee33-4ed3-45d0-823c-6e45cd819789","arxiv_id":"2505.04316","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Every special triple of representations in SU(2) and SU(3) WZW theories contains a simple current, fully classifying the boundary operators invariant under topological defect junctions in these models.","lead":"This paper classifies, for SU(2) and SU(3) WZW conformal field theories, the triples of representations for which a boundary operator can commute with a topological defect junction. It finds that every such triple is built from a simple current, giving an explicit list of commuting boundary operators and junctions in these models.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 4.2 rests on Tables 1–3 and Statements 1–15, for which Appendix B provides only sample proofs; a missed coexistence case or incorrect witness would invalidate the SU(3) classification.","rationale":"The reader's weakest assumption identifies the same load-bearing concern: the SU(3) classification in Theorem 4.2 depends on a case analysis whose completeness is not fully demonstrated. The paper itself flags this by saying that Appendix B gives only sample proofs of some statements, and Section 5 describes the SU(3) proof as brute force. This is a proof-completeness gap rather than a demonstrated mathematical error: the SU(2) theorem is fully proved, the general coefficient formula in Appendix A is a solid contribution, and all presented examples appear consistent. My stress-test found an additional wrinkle in the reduction step: (4.40) as written uses R^k_(a,b),(m,n), but the correct reduction of the four-point vacuum multiplicity requires R^k_(a,b),(n,m). This is probably a repairable typo because the tables include conjugate sets, but it reinforces that the proof of Theorem 4.2 needs a careful, complete rewrite rather than a sample. The appropriate verdict is CONDITIONAL: the classification should be accepted only once Statements 1–15 are fully proved or independently machine-verified, and once the conjugation issue in (4.40) is corrected.","tokens_in":20280,"tokens_out":21869,"duration_ms":204473,"concrete_test":"Implement the BMW formula (4.1) together with the set definitions (4.10)–(4.12), (4.17)–(4.18), and (4.20)–(4.32) in a computer algebra system. For every pair of non-simple-current sets appearing in Tables 1–3, solve the coexistence conditions on (a,b,k) symbolically and verify that the listed witness w_l lies in R^k_(a,b),(i,j) ∩ R^k_(a,b),(n,m), using the conjugate of the column set as required by the vacuum-multiplicity identity. Also run an exhaustive numerical scan over all 0 ≤ b ≤ a ≤ k ≤ 40 with a+b ≤ k to catch any unlisted coexisting pair. The permanent fix is to supply full proofs of Statements 1–15; a single failed entry or missed coexistence would disprove Theorem 4.2.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central SU(3) claim, Theorem 4.2, is proved by a finite case split: (i,j),(m,n) are assigned to the sets in (4.20), and Tables 1–3 give a witness w_l satisfying (4.40). The paper states in §4.4 that Appendix B contains “the minimal set of statements… and a sample proof of some of them”, and Appendix B indeed fully proves only Statements 1 and 2; Statements 3–15 and all coexistence “-” entries are asserted without proof. If any pair of sets can coexist for some k,a,b without the listed witness, or if any restriction indicated in red is incomplete, inequality (4.37) can fail even though every presented example is correct. The completeness of the set decomposition R^(1);k_(a,b),(b,a) in (4.20) is also asserted from solving multiplicity-1 conditions in each T_i, and the six witnesses are claimed to cover all cases without a completeness proof. A second, smaller issue: the reduction to (4.40) should use the conjugate: because the vacuum multiplicity equals Σ_c N^c_{(a,b),(i,j)} N^c_{(a,b),(n,m)}, the second intersection should be R^k_(a,b),(n,m), not R^k_(a,b),(m,n). Since the tables include both a set and its conjugate, this is likely repairable, but as written the proof statement needs correction.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies local fields invariant under the action of topological defects in rational conformal field theories. It derives a general TFT expression for the coefficients governing the sweep of a topological defect past a bulk field, and then reformulates the existence of invariant fields in the charge-conjugation modular invariant case as a fusion-rule condition on 'special triples' (a,x,i). The bulk version of this condition was previously studied; the boundary version is new. The paper proves that for SU(2) WZW theory all special triples come from simple currents, and claims the same classification for SU(3) WZW theory, with an explicit list of all special triples in Section 4.5. The SU(3) proof is based on an explicit description of multiplicity-one representations in the fusion (a,b)x(b,a), followed by a finite case analysis summarized in Tables 1-3 and supported by statements in Appendix B.","tokens_in":20494,"tokens_out":13755,"duration_ms":124149,"significance":"If the SU(3) classification is correct, the paper gives a useful and nontrivial result: it solves a new fusion-rule problem for boundary operators and topological junctions, provides the first complete classification of such 'special triples' for SU(3), and offers a concrete direction for general WZW theories. The general sweep-coefficient formula (1.1) and its derivation in Appendix A are a valuable technical contribution. The SU(2) theorem is cleanly proved, and the explicit description of the SU(3) special triples in Section 4.5 is a concrete payoff. The main weakness is that the central SU(3) theorem rests on a large case analysis whose supporting statements are not fully proved in the manuscript.","major_comments":[{"comment":"Theorem 4.2 and the proof in §4.4 rely on Tables 1-3, which are justified by Statements 1-15 in Appendix B. The appendix gives full proofs only for Statements 1 and 2; Statements 3-15 are asserted without proof, and §4.4 explicitly says the appendix contains 'a sample proof of some of them'. This is a load-bearing gap. The '-' entries in Tables 1-3 assert that certain pairs of sets cannot coexist for given k,a,b, and the completeness of the case split depends on every membership, coexistence, and witness statement. If any pair of sets can coexist without the listed witness, inequality (4.37) can fail even if all displayed examples are correct. Please provide complete proofs for Statements 3-15 and for the non-coexistence entries, or a machine-checkable verification of the tables.","section":"§4.4 and Appendix B"},{"comment":"The reduction in Eq. (4.40) uses the intersection R^k_{(a,b),(i,j)} ∩ R^k_{(a,b),(m,n)}, but the vacuum multiplicity is Σ_c N^c_{(a,b),(i,j)} N^{c*}_{(b,a),(m,n)} = Σ_c N^c_{(a,b),(i,j)} N^c_{(a,b),(n,m)} (using conjugation). The second set should therefore be R^k_{(a,b),(n,m)}, not R^k_{(a,b),(m,n)}. The tables contain conjugate sets, so the intended argument is likely repairable, but the displayed reduction as written is not correct.","section":"§4.4, Eq. (4.40)"},{"comment":"The decomposition of R^{(1);k}_{(a,b),(b,a)} into the sets (4.21)-(4.32) is presented as the result of 'solving' the multiplicity-one conditions, but no derivation is shown for the completeness of this list or for the parameter ranges. This decomposition is the starting point of the exhaustive case split in §4.4. Without a proof that no other multiplicity-one representations occur for arbitrary a,b,k, the subsequent case analysis is not exhaustive. Please add the derivation or a clear reference.","section":"§4.3, Eq. (4.20)"}],"minor_comments":[{"comment":"The notation 'R^{2a}_{a,a}' is nonstandard; it should be R^k_{a,a} with k=2a, since the notation R^k is used throughout the paper.","section":"§3, after Theorem 3.1"},{"comment":"The identification R^{(1);3a}_{(a,a),(a,a)} = W1 ∪ \\bar W1 ∪ W is asserted without derivation; a short verification starting from (4.20)-(4.32) would help the reader trust the explicit list.","section":"§4.5, Eq. (4.43)"},{"comment":"The red/black colour coding in Tables 1-3 and the figures is not legible in grayscale; please add textual markers or hatching to distinguish the cases.","section":"Tables 1-3 and Figures 4-6"},{"comment":"The numerical search for solutions of Eq. (5.1) is described without specifying the range of k,a,b or the method used; as stated it is not reproducible. Either give details or present it as a heuristic remark.","section":"§5, final paragraph"}],"recommendation":"major_revision","confidential_remarks":"The main theorem's proof is incomplete as submitted: Statements 3-15 in Appendix B are asserted but not proved, and the paper itself acknowledges that only a sample proof is provided. This is fixable but requires substantial additional work, either complete proofs or a verified computer-assisted check of Tables 1-3. I would not recommend rejection, as the central claim appears plausible and the SU(2) part is solid, but the manuscript does not currently meet the proof standard for its central result."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"New boundary special-triples problem, clean SU(2) proof, but the SU(3) classification rests on a case analysis that the paper itself only partially proves. That is the thing to know before citing Theorem 4.2.\n\nThe paper identifies a question nobody had addressed: when does a boundary operator and a topological junction leave a one-dimensional invariant subspace, in charge-conjugation RCFTs? The bulk analogue was studied in [1], but the boundary version is genuinely new. Appendix A's derivation of the general sweep coefficient (1.1) is a useful, self-contained piece of TFT bookkeeping. The SU(2) theorem is fully proved and gives a clean result: special triples exist only when a simple current is present. The minimal-model application is a nice bonus.\n\nThe SU(3) part is where I get cautious. The BMW formula is used correctly to describe the multiplicity-one representations in (a,b)×(b,a), and the geometric picture (weights on the sides of a polygon) is informative. But Theorem 4.2 is proved by the tables in Section 4.4, and Appendix B states that only a sample proof of the supporting statements is included. Statements 1 and 2 are proved; Statements 3–15 are asserted. The entries \"–\" (sets that allegedly never coexist) are also asserted. If even one coexistence case is missing, or one witness w_l is not actually in the claimed intersection, the conclusion could fail despite every example shown being correct. The paper says the proof is 'brute force' and leaves a geometric proof to future work, which is honest but confirms the gap.\n\nThere is also a small technical slip in the reduction to (4.40): the second fusion should use the conjugate of (m,n), i.e. R^k_{(a,b),(n,m)} rather than R^k_{(a,b),(m,n)}. The tables are set up symmetrically under conjugation, so this is probably repairable, but as written the proof statement is not correct.\n\nOverall, this is a solid paper with one load-bearing gap. The problem is new, the SU(2) theorem is solid, and the SU(3) classification is plausible and explicit. But the central theorem is not yet fully demonstrated. I would send it to peer review—the referees should ask for the complete proofs of Statements 3–15, or a clear split between proved and conjectural parts. If the gap can be closed, this is a useful contribution to defect CFT. If not, the paper still contains a good new problem and a clean SU(2) solution, but the SU(3) claim needs to be softened. The likely readership is people working on topological defects, RCFT boundaries, and fusion-rule techniques; anyone relying on the SU(3) classification should wait until the appendix is completed.","headline":"New boundary special-triples problem, clean SU(2) proof, but the SU(3) classification hangs on an asserted case analysis that the appendix only partially proves.","tokens_in":21103,"tokens_out":8644,"would_cite":true,"duration_ms":73846,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["81T40","17B67"],"pacs":["11.25.Hf"],"model":"deepseek-v4-flash","headline":"This paper proves that in SU(2) and SU(3) WZW theories, the only triples of representations satisfying the boundary special-triple conditions are those containing a simple current, and it gives the complete explicit list for SU(3).","keywords":["rational conformal field theory","topological defects","WZW models","fusion rules","simple currents","boundary conformal field theory","special triples","SU(3)"],"falsifier":"A brute-force scan over levels $k$ and weights $a\\ge b$ using the BMW formula could settle the claim: list all non-simple-current representations in $R^{(1);k}_{(a,b),(b,a)}$ and test each pair $(i,j),(m,n)$ to see whether $(a,b)\\times(i,j)$ and $(a,b)\\times(m,n)$ share any representation beyond $(a,b)$. A single pair whose only common representation is $(a,b)$ would give $\\dim\\operatorname{Hom}((a,b)\\times(b,a)\\times(i,j)\\times(m,n),(0,0))=1$, contradicting Theorem 4.2 and exposing a gap in the case analysis.","tokens_in":20012,"feed_emoji":"⚛️","tokens_out":13339,"duration_ms":125748,"temperature":0.7,"pith_summary":"This paper asks when a local field in a rational conformal field theory can commute or anticommute with a topological defect, meaning that sweeping the defect past the field produces no extra non-local terms. In theories with the charge-conjugation modular invariant this becomes a question about fusion rules, and for boundary operators the paper isolates a precise condition: a 'special triple' of representations $(a,x,i)$ in which both $x$ and $i$ appear with multiplicity one in $a\\times\\bar a$, and the four-fold fusion with the vacuum has multiplicity exactly one. The paper solves this special-triple problem for the SU(2) and SU(3) Wess\\textendash{}Zumino\\textendash{}Witten models, proving that every special triple must contain a simple current\\textemdash{}a representation whose fusion with any other representation is a single irreducible representation. For SU(3) the complete list is given explicitly: apart from the trivial case where the boundary label is itself a simple current, one must have $a=b=k/3$ and one of the two other labels must be one of the two non-identity simple currents, with the remaining label any multiplicity-one weight on the sides of the triangular fusion polygon. This matters because special triples are the fusion-rule data that produces boundary operators invariant under topological junctions, and such operators constrain boundary renormalization-group flows.","feed_headline":"Simple currents explain all SU(3) special triples","feed_subtitle":"Complete classification shows every special triple of representations includes a simple current.","key_machinery":"The load-bearing object is the special triple: a triple $(a,x,i)$ of representations in which both $x$ and $i$ appear with multiplicity one in $a\\times\\bar a$ and the vacuum appears with multiplicity one in the four-fold fusion $a\\times\\bar a\\times i\\times x$. For SU(3) the calculation is carried by the BMW formula, the explicit Begin\\textendash{}Mathieu\\textendash{}Walton expression for $\\mathrm{bsu}(3)_k$ fusion multiplicities, which organises the multiplicity-one representations in $(a,b)\\times(b,a)$ into disjoint sets lying on the sides of a polygon in the weight lattice. The proof then uses six witness weights $w_1,\\dots,w_6$ near $(a,b)$; the tables in Section 4.4 record that for every pair of non-simple-current multiplicity-one labels at least one witness lies in both fusions $(a,b)\\times(i,j)$ and $(a,b)\\times(m,n)$, which forces the hom space to have dimension greater than one.","core_discovery":"The paper's central claim is Theorem 4.2: let $(a,b)$ be an integrable $\\mathrm{bsu}(3)_k$ representation that is not a simple current, and let $(i,j),(m,n)$ be two representations from $R^{(1);k}_{(a,b),(b,a)}$ that are also not simple currents. Then $\\dim\\operatorname{Hom}((a,b)\\times(b,a)\\times(i,j)\\times(m,n),(0,0))>1$. Because equality to one is the defining condition of a special triple, the theorem forces at least one of the two non-boundary labels to be a simple current. The same conclusion is proved for SU(2) in Theorem 3.1, and the paper derives the minimal-model classification from that SU(2) result. For SU(3), Section 4.5 turns the theorem into a complete list: the only non-trivial case has boundary label $(a,a)$ with $k=3a$, and the triples consist of a simple current together with one of the multiplicity-one weights on the sides of the triangular fusion polygon, as written in equations (4.43) and (4.44).","pith_inferences":["If the simple-current statement survives for other SU(N) groups, as the authors conjecture, then the fusion-rule route to defect-invariant boundary operators will only ever produce operators tied to the center $Z_N$ symmetry, and a geometric intersection proof for the relevant polytopes would settle the conjecture.","Because equations (1.7) and (1.8) are sufficient rather than necessary, this classification does not exclude invariant boundary operators whose non-local coefficients vanish through the actual fusing matrices; a complete enumeration would still require evaluating the $F$-symbols, for example with tube-algebra methods.","The minimal-model result contains a second family of special triples built from the subrings $(r,1)$ and $(1,s)$ of the Kac table, so analogous factorised constructions in coset or orbifold theories may produce non-simple-current special triples even when the WZW parent theory has none."],"forward_implications":["For SU(2) and SU(3) WZW models, every special triple contains a simple current, so the fusion-rule construction yields no invariant boundary configurations outside the simple-current family.","For SU(3) the classification is complete and explicit: the only non-trivial boundary label is $(a,a)$ at level $3a$, and the other two labels are a simple current together with any multiplicity-one weight on the boundary of the triangular fusion polygon.","For minimal models, special triples come in two families: one inherited from the SU(2) simple-current classification and one built from the subrings $(r,1)$ and $(1,s)$ of the Kac table.","The SU(3) proof reduces the classification to showing that certain pairs of representation sets always intersect, which is the natural geometric formulation to generalise to other WZW models."],"supporting_citations":[{"why":"Supplies the explicit bsu(3)_k fusion-coefficient formula used to decompose $(a,b)\\times(b,a)$ and to isolate the multiplicity-one sets.","marker":"[29]"},{"why":"Defines the bulk $a\\times b=c$ fusion-rule problem that motivates the boundary special-triple question and records the prior non-simple-current result for fusion categories.","marker":"[1]"},{"why":"Provides the result used to conclude that in WZW theories the bulk condition has no non-simple-current solutions.","marker":"[21]"},{"why":"Gives the fusing-matrix description of moving a defect-boundary junction past a boundary field, leading to equations (1.7) and (1.8).","marker":"[22]"},{"why":"Supplies the fusion-rule restrictions that determine which intermediate defects survive when a topological defect is swept across a local field.","marker":"[15]"},{"why":"Provides the TFT correlator framework used to derive the general coefficient formula for sweeping a defect past a bulk field.","marker":"[16]"}],"fun_headline_variants":["SU(3) special triples always include a simple current","Simple currents are mandatory in SU(3) special triples","Topological defects force simple currents in SU(3) triples","Every SU(3) special triple contains a simple current","No SU(3) special triple without a simple current"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The SU(3) theorem depends on the completeness of the case analysis summarised in Tables 1 to 3; the appendix proves only a sample of the supporting statements, so if any coexistence region for the multiplicity-one sets was missed, a non-simple-current special triple could exist even though every displayed example is correct.","fun_headline_variants_meta":{"raw":{"variants":["SU(3) special triples always include a simple current","Simple currents are mandatory in SU(3) special triples","Topological defects force simple currents in SU(3) triples","Every SU(3) special triple contains a simple current","No SU(3) special triple without a simple current"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000407,"raw_usage":{"total_tokens":2121,"prompt_tokens":955,"completion_tokens":1166,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":571,"completion_tokens_details":{"reasoning_tokens":1081}},"tokens_in":571,"tokens_out":1166,"duration_ms":9097,"temperature":1.0,"reasoning_tokens":1081,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T23:31:22.704330+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A brute-force scan over levels $k$ and weights $a\\ge b$ using the BMW formula could settle the claim: list all non-simple-current representations in $R^{(1);k}_{(a,b),(b,a)}$ and test each pair $(i,j),(m,n)$ to see whether $(a,b)\\times(i,j)$ and $(a,b)\\times(m,n)$ share any representation beyond $(a,b)$. A single pair whose only common representation is $(a,b)$ would give $\\dim\\operatorname{Hom}((a,b)\\times(b,a)\\times(i,j)\\times(m,n),(0,0))=1$, contradicting Theorem 4.2 and exposing a gap in the case analysis.","supporting_citations":[{"cited_title":"su(3)k fusion coefficients","cited_arxiv_id":"hep-th/9206032","evidence_quote":"Supplies the explicit bsu(3)_k fusion-coefficient formula used to decompose $(a,b)\\times(b,a)$ and to isolate the multiplicity-one sets."},{"cited_title":"$a\\times b=c$ in $2+1$D TQFT","cited_arxiv_id":"2012.14689","evidence_quote":"Defines the bulk $a\\times b=c$ fusion-rule problem that motivates the boundary special-triple question and records the prior non-simple-current result for fusion categories."},{"cited_title":"Urichuk and M","cited_arxiv_id":null,"evidence_quote":"Provides the result used to conclude that in WZW theories the bulk condition has no non-simple-current solutions."}],"review_version":1}