{"id":"343efbc5-545b-4f8e-9456-65c5fee0a5ec","arxiv_id":"2605.21724","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"TBP-mHC proposes parameterizations of the Birkhoff polytope via transportation polytopes that achieve exact double stochasticity for hyper-connections using only (n-1)^2 degrees of freedom.","lead":"This paper introduces Transportation Birkhoff Polytope (TBP) and Recursive TBP (RTBP) parameterizations to create exactly doubly stochastic mixing matrices for hyper-connections in residual networks. The goal is to improve training stability and scalability in language models while keeping full expressivity without iterative normalization or high combinatorial costs.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.3","headline":"TBP/RTBP parameterization may span only a submanifold of the Birkhoff polytope rather than the full set for n used in LM experiments","rationale":"The reader's weakest_assumption exactly isolates the surjectivity question that must hold for the headline claim to be true. Because the original review was abstract-only and the full derivations are now available, the concrete numerical check above directly tests whether the claimed full expressivity materializes for the relevant matrix sizes; a negative result would force the verdict from UNVERDICTED toward CONDITIONAL or REJECT.","tokens_in":1680,"tokens_out":392,"duration_ms":35458,"concrete_test":"For n=4 (a plausible size for hyper-connection mixing), generate 50 random doubly stochastic matrices via Sinkhorn on uniform-[0,1] seeds. For each target M, solve the unconstrained optimization over the (n-1)^2=9 TBP parameters to minimize ||TBP(params) - M||_F; if any matrix yields residual >1e-8 after convergence, the parameterization does not achieve full expressivity.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim requires that the TBP construction (and its recursive RTBP variant) is surjective onto the entire Birkhoff polytope of n×n doubly stochastic matrices. While the parameter count (n-1)^2 matches the known dimension, surjectivity is not automatic: the transportation-polytope encoding could restrict the image if the free variables are mapped through a non-bijective or inequality-constrained transformation that excludes some interior or boundary points, especially once the recursive Kronecker-style construction is applied. The abstract asserts “full expressivity” and “exactly doubly stochastic” matrices without iterative normalization, but provides no explicit inverse map or proof that every doubly stochastic matrix is attainable for the concrete n appearing in the language-model runs.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The paper proposes Transportation Birkhoff Polytope (TBP) parameterizations and their recursive variants (RTBP) for manifold-constrained hyper-connections (mHC). It claims these construct exactly doubly stochastic mixing matrices with exactly (n-1)^2 degrees of freedom, achieve full expressivity over the Birkhoff polytope, avoid iterative normalization and combinatorial explosion, and yield competitive performance with improved stability in language-model pre-training.","tokens_in":1846,"tokens_out":592,"duration_ms":27568,"significance":"If the surjectivity claim holds, the work would supply a practical, exact parameterization of the full Birkhoff polytope that matches its known dimension while remaining differentiable and free of iterative projections. This would directly address the expressivity–scalability trade-off left open by Sinkhorn-based mHC and permutation-based mHC-lite, and could be adopted in any architecture that requires learnable doubly stochastic mixing.","major_comments":[{"comment":"§3 (TBP construction) and §4 (RTBP): the central claim that the parameterization is surjective onto the entire Birkhoff polytope is load-bearing for the title and abstract. The parameter count (n-1)^2 matches the dimension, yet surjectivity is not automatic for a transportation-polytope encoding; an explicit inverse map, a density argument, or a constructive proof that every interior and boundary point is attainable must be supplied. Without it the “full expressivity” assertion remains an unverified assertion rather than a theorem.","section":"§3 and §4"},{"comment":"Experimental section (language-model runs): the reported n values and the concrete mixing matrices used in the LM experiments should be checked against the claimed surjectivity. If the recursive Kronecker-style construction in RTBP introduces additional linear dependencies, the effective image may be a proper submanifold for the n appearing in the tables; an ablation that samples random doubly stochastic targets and measures reconstruction error under the learned parameterization is required to substantiate the claim.","section":"Experimental section"}],"minor_comments":[{"comment":"Abstract: stray apostrophe in “pre-training'” should be removed.","section":"Abstract"},{"comment":"Notation: define the precise mapping from free variables to the transportation polytope entries (e.g., the role of the marginal vectors) before the recursive construction is introduced; the current presentation leaves the base TBP map implicit.","section":"§3"}],"recommendation":"major_revision","confidential_remarks":"The manuscript cites the transportation-polytope literature but does not discuss whether the proposed encoding is equivalent to any existing parameterization of the Birkhoff polytope (e.g., via Birkhoff–von Neumann or other convex-hull representations). A short comparison paragraph would strengthen the novelty claim."},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the careful and constructive review. The comments highlight important points regarding the rigor of our surjectivity claim and the need for additional empirical validation. We address each major comment below and will incorporate revisions to strengthen the manuscript.","responses":[{"response":"We agree that a formal proof of surjectivity is necessary to substantiate the central claim. In the revised manuscript we will add an explicit constructive proof in a new subsection of §3. For any target doubly stochastic matrix B, we exhibit a closed-form inverse that recovers the (n-1)×(n-1) transportation parameters whose associated transportation polytope projects exactly onto B; the construction handles both interior points and all permutation-matrix boundary points. For the recursive RTBP construction in §4 we will prove by induction that surjectivity is preserved and that no additional linear dependencies arise, so the image remains the full Birkhoff polytope with exactly (n-1)^2 degrees of freedom.","revision_made":"yes","referee_comment":"[§3 and §4] §3 (TBP construction) and §4 (RTBP): the central claim that the parameterization is surjective onto the entire Birkhoff polytope is load-bearing for the title and abstract. The parameter count (n-1)^2 matches the dimension, yet surjectivity is not automatic for a transportation-polytope encoding; an explicit inverse map, a density argument, or a constructive proof that every interior and boundary point is attainable must be supplied. Without it the “full expressivity” assertion remains an unverified assertion rather than a theorem."},{"response":"We will add the requested ablation study to the experimental section. For each n appearing in the language-model tables we will sample several thousand random doubly stochastic matrices (including boundary permutations) and report the reconstruction error obtained by applying the inverse TBP/RTBP map. We will also explicitly state the n values used and confirm that the mixing matrices realized during pre-training lie in the attainable set. This empirical check will be presented alongside the existing results.","revision_made":"yes","referee_comment":"[Experimental section] Experimental section (language-model runs): the reported n values and the concrete mixing matrices used in the LM experiments should be checked against the claimed surjectivity. If the recursive Kronecker-style construction in RTBP introduces additional linear dependencies, the effective image may be a proper submanifold for the n appearing in the tables; an ablation that samples random doubly stochastic targets and measures reconstruction error under the learned parameterization is required to substantiate the claim."}],"tokens_in":1382,"tokens_out":532,"duration_ms":22476,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The main thing to know is that the paper introduces transportation-polytope (TBP) and recursive (RTBP) parameterizations that produce exactly doubly stochastic mixing matrices for hyper-connections using exactly (n-1)^2 free parameters. This sits between the approximate Sinkhorn approach in mHC and the factorial cost of mHC-lite, while trying to remove the submanifold restriction that KromHC imposes. The recursive construction is the practical piece that could scale to the matrix sizes appearing in residual streams for language models. The pre-training experiments report competitive performance with gains in stability and scalability, which is the result that would matter if the parameterization works as stated. The soft spot is the surjectivity question. Matching the dimension of the Birkhoff polytope is necessary but not sufficient; the map from free variables through the transportation constraints could still miss some matrices, especially after the recursive Kronecker-style layering. The abstract asserts full expressivity without showing an inverse construction or even a numerical check that recovers arbitrary target matrices for small n. If the full text contains a clear proof or such a verification, that would close the gap. Otherwise the central claim rests more on the parameter count than on demonstrated reachability. This is for people already working on manifold-constrained mixing in residuals or on polytope parameterizations in deep networks. A reader following the mHC line would find the direct comparison useful and might want to plug the parameterization into their own code. It deserves a serious referee because the motivation is concrete, the empirical signal is positive, and the idea is specific enough that referees can check the math and the small-n cases directly.","headline":"TBP/RTBP gives a clean parameterization for exactly doubly stochastic hyper-connections with the right parameter count, but the claim of full expressivity onto the Birkhoff polytope still needs an explicit inverse or small-n verification to be convincing.","tokens_in":2341,"tokens_out":419,"would_cite":false,"duration_ms":32391,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":{"model":"grok-4.3","evidence":[],"headline":"TBP/RTBP parameterization of Birkhoff polytope via North-West Corner transport has no structural overlap with RS forcing chain or J-cost","alignment":"orthogonal","rationale":"The paper's core contribution is an exact, minimal (n-1)^2-parameter chart for the interior of the Birkhoff polytope (doubly-stochastic matrices) realized by sequential budget updates on transportation margins, together with its recursive block decomposition. This machinery lives entirely in the domain of convex geometry and neural-network mixing operators; it invokes neither the recognition cost J, ratio symmetry, golden-ratio identities, 8-tick periodicity, nor any parameter-free derivation of physical constants. No RS theorem (e.g., reality_from_one_distinction, washburn_uniqueness_aczel, alexander_duality_circle_linking, or any Cost/FunctionalEquation result) is paralleled or contradicted.","tokens_in":56347,"confidence":"high","tokens_out":198,"duration_ms":11250,"cache_read_input_tokens":38528,"cache_creation_input_tokens":0},"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"Transportation polytope parameterizations produce exactly doubly stochastic mixing matrices for hyper-connections with only (n-1)^2 free parameters.","keywords":["hyper-connections","doubly stochastic matrices","Birkhoff polytope","transportation polytopes","residual networks","language model pre-training","manifold constraints"],"falsifier":"Exhibiting, for any tested matrix size n, at least one doubly stochastic matrix that cannot be exactly recovered from the TBP map.","tokens_in":2568,"feed_emoji":"🔄","tokens_out":591,"duration_ms":27918,"temperature":0.7,"pith_summary":"Hyper-connections let residual networks mix multiple streams with learned weights, but free mixing often destabilizes training. Earlier fixes either approximate double stochasticity through iterative Sinkhorn steps or enforce it exactly via permutations at factorial cost or via restricted Kronecker structures. The paper replaces those with Transportation Birkhoff Polytope and recursive variants that directly output exact doubly stochastic matrices. The construction uses exactly the dimension of the Birkhoff polytope and skips both normalization loops and combinatorial enumeration. Experiments on language-model pre-training report competitive accuracy together with gains in training stability and scaling behavior.","feed_headline":"Exact doubly stochastic mixes via transportation polytopes","feed_subtitle":"TBP and RTBP hit every matrix in the Birkhoff polytope with (n-1)^2 parameters, removing both Sinkhorn loops and permutation explosion.","key_machinery":"Transportation Birkhoff Polytope (TBP) parameterization, which maps unconstrained parameters onto the full set of doubly stochastic matrices via transportation polytopes.","core_discovery":"TBP and RTBP parameterizations construct exactly doubly stochastic mixing matrices with (n-1)^2 degrees of freedom. The approach avoids iterative normalization and combinatorial explosion while preserving full expressivity of the Birkhoff polytope.","pith_inferences":["The same parameterization could be substituted into any architecture that already imposes manifold constraints on weight matrices.","If the span is complete, gradient flow on the mixing weights should avoid the projection steps that sometimes slow Sinkhorn-based variants.","Direct comparison of the learned mixing matrices against those produced by KromHC would quantify how much additional expressivity is actually used."],"forward_implications":["Mixing matrices satisfy exact double stochasticity at every forward pass without Sinkhorn iterations.","The number of trainable parameters per mixing matrix matches the intrinsic dimension of the Birkhoff polytope.","Complexity remains polynomial in n instead of factorial.","Language-model pre-training reaches competitive performance while showing improved stability and scalability."],"fun_headline_variants":["TBP achieves exact doubly stochasticity with (n-1)^2 degrees of freedom","Full Birkhoff expressivity without iterative normalization using TBP","RTBP parameterizes mixing matrices to avoid permutation explosion","TBP RTBP cover every matrix in the Birkhoff polytope exactly"],"cache_read_input_tokens":64,"weakest_assumption_plain":"The chosen parameterization is assumed to reach every doubly stochastic matrix rather than only a lower-dimensional subset of them.","fun_headline_variants_meta":{"raw":{"variants":["TBP achieves exact doubly stochasticity with (n-1)^2 degrees of freedom","Full Birkhoff expressivity without iterative normalization using TBP","RTBP parameterizes mixing matrices to avoid permutation explosion","TBP RTBP cover every matrix in the Birkhoff polytope exactly"]},"model":"grok-4.3","cost_usd":0.014019,"raw_usage":{"total_tokens":5927,"prompt_tokens":585,"num_sources_used":0,"completion_tokens":73,"cost_in_usd_ticks":140190500,"prompt_tokens_details":{"text_tokens":585,"audio_tokens":0,"image_tokens":0,"cached_tokens":64},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":5269,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":585,"tokens_out":73,"duration_ms":39342,"temperature":1.0,"reasoning_tokens":5269,"cache_read_input_tokens":64,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-05-22T09:56:44.831222+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"Exhibiting, for any tested matrix size n, at least one doubly stochastic matrix that cannot be exactly recovered from the TBP map.","supporting_citations":[],"review_version":1}