{"id":"a64edff9-28e9-4d6e-8c65-a048267181f0","arxiv_id":"2606.01883","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":7.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":1,"one_line_summary":"Balanced equal-norm prototype codes enable theoretical guarantees for simplex-ratio OSR in all embedding dimensions, with a dichotomy at d >= C-1 and exponential decay of false acceptance.","lead":"The paper develops a theoretical analysis of open-set recognition using balanced prototype geometries that applies to any embedding dimension. This framework could improve understanding and design of classifiers that reliably reject unknown inputs in critical applications.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.3","headline":"Theory applies only to balanced equal-norm codes (equal lengths, zero sum); if learned prototypes deviate, ball sublevel sets, dichotomy at d=C-1, and exponential decay do not transfer.","rationale":"The reader's weakest_assumption exactly identifies the load-bearing gap between the proven geometry and its use as a prior. The paper's own empirical caveat (raw ratio scores underperform other rules) reinforces rather than resolves this transfer issue, so the UNVERDICTED verdict requires no adjustment.","tokens_in":1865,"tokens_out":368,"duration_ms":23191,"concrete_test":"On the final prototypes from the CIFAR/MedMNIST experiments, compute max norm deviation across classes and the Euclidean norm of their vector sum; evaluate the resulting defect parameter against the explicit bound in the dichotomy statement. If the observed defect places the configuration outside the controlled-degradation regime, the ball-union and symmetry claims do not hold for the reported runs.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The strongest claim centers on mathematical properties proven specifically for balanced equal-norm codes: auxiliary squared-ratio sublevel sets are exact unions of Euclidean balls that bracket the operational score's acceptance region; a sharp if-and-only-if dichotomy on one-distance symmetry holds exactly when d >= C-1 (with explicit defect parameter controlling degradation below the threshold); and false-acceptance rate decays exponentially under isotropy assumptions. The paper treats these codes as both analytic tool and representation-learning prior rather than claiming they arise automatically in deployed models. No quantitative verification is described (in the provided abstract) of how closely optimized embeddings satisfy equal-norm and zero-sum conditions, nor of the realized defect parameter. If deviation is large, the bracketing, symmetry, and decay results cease to apply to the operational system.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The paper presents a theoretical analysis of open-set recognition (OSR) using balanced equal-norm prototype codes that extends simplex-based methods to embedding dimensions d < C-1. It proves that for these codes, the sublevel sets of an auxiliary squared ratio score are exact unions of Euclidean balls that bracket the acceptance region of the operational score. A sharp dichotomy is established: one-distance symmetry holds if and only if d >= C-1, with an explicit defect parameter governing degradation otherwise. Additional results include exponential decay of the false-acceptance rate under isotropy assumptions and the global Lipschitz continuity of the operational score with compact acceptance regions. Empirically, the geometry is evaluated as an analytic tool and representation-learning prior on CIFAR and MedMNIST datasets, where it provides useful structure but OSR performance is shown to depend strongly on the scoring rule, with raw ratio scores underperforming nearest-neighbor and logit-based alternatives.","tokens_in":2056,"tokens_out":482,"duration_ms":25251,"significance":"If the theoretical results hold, this work provides a rigorous geometric foundation for prototype-based OSR that applies in all dimensions, including the previously unanalyzed regime d < C-1. The proofs of the ball sublevel sets, the if-and-only-if symmetry dichotomy, and the exponential decay guarantee represent significant contributions to the theoretical understanding of OSR. The explicit defect parameter offers a concrete way to quantify performance degradation in low dimensions. The empirical component, while not claiming state-of-the-art performance, demonstrates the geometry's utility as a prior. These elements strengthen the manuscript's value for the field of open-set recognition in safety-critical applications like medical imaging.","major_comments":[{"comment":"Abstract and theoretical analysis: The proofs of ball sublevel sets, the symmetry dichotomy, and exponential decay are derived specifically for exact balanced equal-norm codes (equal lengths, zero sum). The manuscript does not report quantitative verification (e.g., measured norm deviations, zero-sum residuals, or realized defect parameter values) of how closely the learned prototypes satisfy these conditions in the CIFAR and MedMNIST experiments. This is load-bearing, as substantial deviation would mean the bracketing property and other guarantees do not transfer to the operational systems evaluated.","section":"Abstract / Empirical Evaluation"}],"minor_comments":[],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the thorough review and for recognizing the theoretical contributions on balanced equal-norm codes, the ball-sublevel-set property, the symmetry dichotomy, and exponential decay. The single major comment identifies a genuine gap in bridging the exact-code theory to the learned prototypes in experiments. We address it directly below and will incorporate the requested verification.","responses":[{"response":"We agree that the theoretical results are stated for exact balanced equal-norm codes and that the manuscript currently lacks quantitative checks on how closely the learned prototypes match these conditions. This is a valid concern for assessing transfer of the bracketing and symmetry guarantees. In the revised version we will add explicit measurements—norm deviations, zero-sum residuals, and realized defect-parameter values—for the prototypes obtained on both CIFAR and MedMNIST splits, together with a brief discussion of their implications for the empirical results.","revision_made":"yes","referee_comment":"[Abstract / Empirical Evaluation] Abstract and theoretical analysis: The proofs of ball sublevel sets, the symmetry dichotomy, and exponential decay are derived specifically for exact balanced equal-norm codes (equal lengths, zero sum). The manuscript does not report quantitative verification (e.g., measured norm deviations, zero-sum residuals, or realized defect parameter values) of how closely the learned prototypes satisfy these conditions in the CIFAR and MedMNIST experiments. This is load-bearing, as substantial deviation would mean the bracketing property and other guarantees do not transfer to the operational systems evaluated."}],"tokens_in":1581,"tokens_out":322,"duration_ms":12094,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The main takeaway is that this paper gives a geometric account of open-set recognition that works in every embedding dimension by shifting from regular simplices to balanced equal-norm codes (equal lengths, zero sum). These codes exist for d at least 2 and recover the simplex when d is large enough.\n\nWhat is new is the set of properties they derive for these codes: an auxiliary squared-ratio score has sublevel sets that are exact unions of Euclidean balls bracketing the operational acceptance region; a sharp if-and-only-if dichotomy on one-distance symmetry that holds precisely when d is at least C-1, with an explicit defect parameter for lower dimensions; plus exponential decay of false-acceptance rate under isotropy and global Lipschitz continuity with compact regions. The abstract presents these as formal results rather than heuristics.\n\nThe work does a reasonable job treating the codes as both an analytic device and a representation-learning prior, and the empirical section on CIFAR and MedMNIST is honest about its limits: the geometry adds structure but raw ratio scores underperform nearest-neighbor and logit baselines.\n\nThe soft spot is the transfer question. The derivations assume prototypes stay close to equal-norm and zero-sum. The paper does not report any quantitative check on how well optimized embeddings satisfy those conditions or what defect parameter is realized in practice. If the match is loose, the ball bracketing, dichotomy, and decay guarantees do not carry over to the deployed system.\n\nThis is for readers who care about the geometry of OSR or embedding constraints in low dimensions. Practitioners hunting for stronger detectors will find little to use. The theory is grounded enough to merit a serious referee even with the modest empirical claims.","headline":"Extends simplex OSR theory to any dimension via balanced equal-norm codes with clean proofs on ball sublevel sets and a symmetry dichotomy, but the practical value hinges on an unverified assumption about learned prototypes.","tokens_in":2518,"tokens_out":424,"would_cite":false,"duration_ms":14200,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"Balanced equal-norm codes extend ratio-based open-set recognition to every embedding dimension.","keywords":["open-set recognition","prototype geometry","balanced codes","ratio score","embedding dimension","simplex","false acceptance rate","Lipschitz continuity"],"falsifier":"Checking whether the sublevel sets of the squared ratio score remain exact unions of Euclidean balls when prototypes violate equal norms or zero-sum in dimensions below C-1.","tokens_in":2767,"feed_emoji":"📐","tokens_out":658,"duration_ms":19232,"temperature":0.7,"pith_summary":"The paper establishes a theoretical account of simplex-ratio open-set recognition that holds for all embedding dimensions, not only those high enough to contain a regular simplex. It centers analysis on balanced equal-norm codes, configurations of class prototypes with equal lengths and zero sum that exist for any d at least 2. This matters because it supplies guarantees on acceptance regions and error rates that apply in the low-dimensional regimes typical of practical embeddings. The account proves that an auxiliary squared ratio score has sublevel sets that are exact unions of Euclidean balls bracketing the operational score, together with a sharp condition on when the prototypes recover simplex-like symmetry.","feed_headline":"Balanced codes let ratio scores work in any embedding dimension","feed_subtitle":"Squared-ratio sublevel sets form ball unions that bracket acceptance regions, with exponential false-accept decay in d.","key_machinery":"balanced equal-norm codes (equal-length, zero-sum prototypes) whose induced squared-ratio sublevel sets are exact unions of Euclidean balls","core_discovery":"For balanced equal-norm codes the squared ratio score has sublevel sets that are exact unions of Euclidean balls bracketing the acceptance region of the operational score. The prototypes attain one-distance symmetry if and only if the embedding dimension d is at least C minus one, with controlled degradation governed by an explicit defect parameter below that threshold. Under natural isotropy assumptions the false-acceptance rate decays exponentially in d, and the operational score is globally Lipschitz with compact acceptance regions.","pith_inferences":["The defect parameter below the simplex threshold could be used to adjust scoring rules in a dimension-dependent way.","Treating balanced geometry as a soft constraint during embedding training might improve robustness of downstream open-set detectors.","Ball-union bracketing may permit faster verification of acceptance regions by reducing the problem to distance checks against prototype centers.","The same geometry might transfer to other distance-ratio methods in classification or anomaly detection that rely on prototype symmetry."],"forward_implications":["Sublevel sets of the squared ratio score are exact unions of Euclidean balls that bracket the operational acceptance region.","Prototypes achieve one-distance symmetry precisely when d is at least C-1, otherwise governed by an explicit defect parameter.","False-acceptance rate decays exponentially in embedding dimension under isotropy assumptions.","The operational score is globally Lipschitz continuous and produces compact acceptance regions."],"fun_headline_variants":["Balanced codes support ratio OSR in any embedding dimension","Balanced prototypes extend OSR beyond regular simplex limits","Equal-norm codes enable ratio scores for all embedding dimensions","Theory covers balanced code OSR with ball-union score regions"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"Class prototypes can be configured or learned as balanced equal-norm codes with equal lengths and zero sum.","fun_headline_variants_meta":{"raw":{"variants":["Balanced codes support ratio OSR in any embedding dimension","Balanced prototypes extend OSR beyond regular simplex limits","Equal-norm codes enable ratio scores for all embedding dimensions","Theory covers balanced code OSR with ball-union score regions"]},"model":"grok-4.3","cost_usd":0.005105,"raw_usage":{"total_tokens":2538,"prompt_tokens":777,"num_sources_used":0,"completion_tokens":61,"cost_in_usd_ticks":51049500,"prompt_tokens_details":{"text_tokens":777,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":1700,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":777,"tokens_out":61,"duration_ms":12268,"temperature":1.0,"reasoning_tokens":1700,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-06-28T15:18:24.205001+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"Checking whether the sublevel sets of the squared ratio score remain exact unions of Euclidean balls when prototypes violate equal norms or zero-sum in dimensions below C-1.","supporting_citations":[],"review_version":1}