{"id":"75708e54-41f6-4ac0-812f-682458d20a48","arxiv_id":"2501.08191","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"A classification framework using inner/outer division, Jupiter presence, and gap spacing categorizes nearly all exoplanet systems with three or more planets.","lead":"This paper sorts the known multi-planet systems into a few simple categories, such as tightly packed 'peas in a pod' systems and 'warm Jupiter' systems, using public exoplanet data. The classification gives astronomers a common language to describe system architectures and to test planet formation models.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 130-day minimum outer-planet period is not independently derived, so the ~97% classification rate is partly a self-fulfilling consequence of a catalog-fitted threshold; this should be tested on an independent or corrected sample before the framework is accepted as robust.","rationale":"The reader's weakest assumption is exactly the arbitrariness of the 130-day and period-ratio-5 thresholds, and the paper's own Section 5.2 admits they are over-fit. My stress-test concurs that this is the most load-bearing issue: the headline 'three questions classify ~97%' is a quantitative claim, and its plausibility hinges on whether the threshold is a physically motivated boundary or a catalog-fitting convenience. The paper provides some independent support for the period-ratio-5 value (Mars-Jupiter ratio 6.3, binary-star stability results around 3.3-3.7, asteroid-belt 4:1 resonance argument), but the 130-day value is supported only by a catalog-empirical observation that <100-day Jupiters have nearby companions while >200-day ones do not. The concern is not that the authors are dishonest; they are explicit about the arbitrariness. The problem is that the quantitative strength of the claim ('sufficient to classify ~97% with minimal ambiguity') is not qualified in the abstract or introduction by the dependence on the fitted threshold, and no sensitivity analysis is provided. This justifies leaving the verdict at CONDITIONAL: the framework is plausible and useful, but the central quantitative claim needs an explicit robustness test and data release before it can be accepted as stated. A held-out test against future discoveries is also relevant, but a cutoff-sensitivity analysis on the current catalog is the immediate, feasible check that would settle whether the 97% number is meaningful rather than an artifact of the 130-day choice.","tokens_in":32081,"tokens_out":1323,"duration_ms":13219,"concrete_test":"Recompute the full classification (Tables 2-5 and the 97% unambiguity fraction) using the same publicly available NASA Exoplanet Archive data but with the minimum outer-planet period varied systematically over 80-200 days (and, equivalently, period-ratio cutoff varied over 4-6), reporting the fraction of systems that change class and the change in the ambiguity count. If the 97% claim is stable at the level of a few percent across the entire range, the concern is mitigated; if it shifts by more than ~10%, the headline claim is tuned to the adopted threshold.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim, that three questions classify ~97% of N>=3 systems 'with minimal ambiguity', depends on the specific definition of the inner/outer divide. Section 5.2 defines the divide using a minimum outer-planet period of 130 days, chosen because, in the September 2024 catalog, essentially all Jupiters with nearby low-mass inner companions have periods <130 days (longer-period exceptions Kepler-90 and HD 10180 are then excluded as 'outliers'). The paper itself states that this choice is 'both arbitrary and over-fit to our small sample' (Section 5.2). This creates a circularity: the divide is tuned so that nearly all systems have a clear inner/outer structure, and the 97% claim is then quoted as evidence that the three-question framework is broadly applicable. If the cutoff is moved to 80 or 200 days (as the authors suggest is permissible), the classification of several systems changes (e.g., HD 33142, HD 141399, Kepler-148), and the reported prevalence numbers and the '~97%' unambiguity fraction shift. The paper provides no analysis of how sensitive the 97% figure and the class fractions are to the adopted cutoff, and no released code or filtered catalog to allow an independent check. The published claim is therefore narrower than stated: it establishes that a 130-day cutoff, fitted to the same catalog, can describe most systems, not that the taxonomy would survive a change in threshold or a larger sample.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper proposes an architecture classification for exoplanetary systems based on the September 2024 NASA Exoplanet Archive catalog. Systems are first divided into inner and outer regimes using a period-ratio gap >5 and a minimum outer-planet period of 130 days; inner systems are then classified as peas-in-a-pod, warm Jupiter, closely spaced, or gapped, with further subdivision by gap location. The paper reports that three questions (inner/outer divide, presence of inner Jupiters, and inner gaps with period ratio >5) classify roughly 97% of N>=3 systems with minimal ambiguity, and it gives qualitative prevalence statistics, outlier case studies, and predictions for future classes. It concludes that exoplanetary systems are more uniform than often claimed and that the Solar System fits as a peas-in-a-pod system with an outer giant population.","tokens_in":32500,"tokens_out":3580,"duration_ms":36912,"significance":"If the central claim is robust, the framework would be a useful organizing taxonomy for the full confirmed multiplanet population, complementing earlier work on peas-in-a-pod systems (Weiss et al.) and gap complexity (Gilbert & Fabrycky). The paper is commendably transparent about its qualitative statistical approach and selection effects, and the census tables and architecture figures are valuable resources. However, the headline quantitative claim of ~97% classification with minimal ambiguity is not yet adequately supported: the 130-day minimum outer-planet period is explicitly admitted to be arbitrary and over-fit to the same catalog used to compute the classification rate, and no sensitivity analysis is provided. The paper also introduces several empirical thresholds (e.g., the 1/7 strongly-inverted mass ratio) without independent validation. These issues are load-bearing for the central claim, so the manuscript requires substantive revision.","major_comments":[{"comment":"The central claim that three questions classify ~97% of N>=3 systems depends critically on the 130-day minimum outer-planet period, which is chosen by inspecting the same September 2024 catalog. The paper explicitly states that this definition is 'both arbitrary and over-fit to our small sample' (Section 5.2). Since systems such as HD 33142, HD 141399, and Kepler-148 are discussed as borderline cases, moving the cutoff to 80 or 200 days would change individual classifications and shift the prevalence fractions in Tables 3 and 4. The authors should provide a sensitivity analysis that recomputes the classification counts and the 'unambiguity' fraction over a plausible range of cutoff values (e.g., 80, 100, 160, 200 days), and they should state how much the reported numbers change.","section":"Section 5.2 and Section 1"},{"comment":"The strongly-inverted mass ratio threshold of M_outer/M_inner < 1/7 is defined as 'an empirical result based on looking at the distribution of mass ratios' (Section 6.4). No independent sample, stability test, or goodness-of-fit justification is given for this specific value. The counts of strongly-inverted systems (8 peas-in-a-pod and 8 warm Jupiter systems in Table 5) and the claim that these systems form a 'distinctive dynamical cluster' depend directly on this threshold. The authors should test the sensitivity of these counts to reasonable changes in the threshold (e.g., 1/5 and 1/10) or derive the threshold from a principled statistical procedure.","section":"Section 6.4 and Table 5"},{"comment":"The TTV mass corrections introduce several free parameters: the rocky-planet scaling factor 8.0/5.5, the 1.75 R_Earth boundary, and the pure-iron-core model with a core radius 0.5 R_Earth smaller than the observed radius. These corrections can change whether a planet is classified as a Jupiter, which directly affects the warm-Jupiter vs. peas-in-a-pod counts that are central to the classification. The paper does not report how many planets were affected by each correction, nor how the final classification changes if these parameters are varied within plausible bounds. At minimum, the authors should report the number of affected planets and a brief stability check of the key class fractions.","section":"Section 2.2 and Tables 3, 5"},{"comment":"The 'minimal ambiguity' claim is not quantified. The paper states that only 9 out of 314 N>=3 systems are difficult to classify, which is ~2.9%, but there is no measure of how many systems lie near the boundaries of the definitions (e.g., period ratios near 5, periods near 130 days, or planet radii/masses near the Jupiter cutoff). A robustness measure, such as the number of systems whose classification changes under small perturbations of the thresholds or under plausible measurement uncertainties, would be needed to substantiate the ~97% figure and to make the claim falsifiable.","section":"Section 8.1 and Table 1"}],"minor_comments":[{"comment":"The first paragraph contains a duplicated phrase: 'in our analysis, in our analysis'.","section":"Section 5.1"},{"comment":"The introductory paragraph contains a typo: 'better-charaterized' should be 'better-characterized'.","section":"Section 6"},{"comment":"The phrase 'the gapped systems compromise only a small minority' should use 'comprise' rather than 'compromise'.","section":"Section 5.3"},{"comment":"The caption begins with 'F ramework summary' with an extra space; the same spacing issue appears in several table captions ('T able').","section":"Figure 1 caption"},{"comment":"The abstract quotes ~97%, and Section 8.1 reports 9 out of 314 systems as difficult to classify; the arithmetic (305/314 = 97.1%) should be made explicit so the reader can verify the claim.","section":"Abstract and Section 8.1"}],"recommendation":"major_revision","confidential_remarks":"The paper is honest about its limitations, but the headline 97% classification claim is currently a best-case statistic for thresholds fitted to the same catalog. I would ask the authors to include the sensitivity analysis described in the major comments and, ideally, to release the filtered catalog and classification code so that the community can independently reproduce the census. Without those additions, the published claim is narrower than the abstract suggests, and the framework's robustness to future data remains untested. The paper is a good fit for the journal's scope, and the revision required is substantial but feasible."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Worth your time. Howe et al. build a classification scheme for exoplanet system architectures from the full Exoplanet Archive, and it is the first to combine mass content with period spacing over the whole confirmed population. The three-question framework (inner/outer divide, presence of inner Jupiters, gaps with period ratio >5) plus the warm Jupiter and gap-location subtypes is a natural organizing device, and the census tables and architecture figures are a real resource. The paper is also honest: it explicitly tells you the statistics are qualitative, the sample is biased, and several thresholds are empirical rather than derived.\n\nThe soft spot is the one the stress-test identifies. The 130-day minimum outer planet period is chosen so that nearly all Jupiters with inner companions fall on one side, and the authors say in Section 5.2 that this choice is “both arbitrary and over-fit to our small sample.” That means the ~97% coverage figure is partly a statement about the fitted cutoff, not an independent discovery about nature. They do not test sensitivity to moving the cutoff to 80 or 200 days, and they release neither the filtered catalog nor code, so an independent check is not possible. The same caveat applies to the 1/7 mass-ratio threshold for strongly-inverted pairs, which is admittedly empirical. These are real limits, but they are limits the authors themselves disclose, and they do not invalidate the central qualitative picture. The existence of a peas-in-a-pod majority, a warm Jupiter minority, and a distinguishable inner/outer gap structure is robust to the exact cutoff.\n\nMy main complaint is that the paper could have done more of the work that would make the taxonomy a standard tool rather than a plausible one. A sensitivity analysis of class fractions versus cutoff, or a release of the curated sample, would have turned a conditional result into a reproducible one.\n\nWho benefits: exoplanet demographics people, formation modelers, and anyone designing simulated populations for future observations. It deserves a serious referee. I would send it to review, with the recommendation that the authors release their filtered planetary systems table and add a cutoff-sensitivity table. That is a routine revision, not a conceptual fix. My own verdict is conditional acceptance rather than rejection.","headline":"A genuinely useful taxonomy for exoplanet system architectures, with an honestly flagged overfit threshold that should temper the headline 97% coverage claim until it is tested on held-out data.","tokens_in":32954,"tokens_out":1105,"would_cite":true,"duration_ms":14396,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Three architecture questions classify nearly all multiplanet systems.","keywords":["exoplanet system architectures","classification scheme","peas in a pod systems","warm Jupiter systems","inner and outer planet divide","gapped systems","NASA Exoplanet Archive","multiplanet systems"],"falsifier":"Re-run the classification on the same 314 systems with the inner/outer cutoff moved to 80 days and 200 days and the gap threshold moved to 4 and 6; the framework's claim would fail if many systems change class or if the fraction classified “with minimal ambiguity” drops well below the reported 97%. A second decisive test is the discovery of a non-Jovian outer planet, which the framework predicts should be rare enough to leave a conspicuous empty class.","tokens_in":31827,"feed_emoji":"🪐","tokens_out":5049,"duration_ms":46572,"temperature":0.7,"pith_summary":"This paper tries to show that the bewildering variety of known planetary systems is actually organized into a small number of architecture types, and that three simple questions are enough to sort nearly all of them. Working from the complete confirmed-exoplanet catalog, the authors split each system into inner and outer planets, ask whether the inner planets include a Jupiter, and ask whether the inner planets contain a gap with a period ratio larger than 5. These three questions classify about 97 percent of systems with three or more planets, with the rest mostly fitting into named subcategories such as “gapped” and “warm Jupiter” systems. If the claim holds, exoplanetary systems are far more uniform than the “zoo” narrative suggests, and the Solar System is a fairly ordinary peas-in-a-pod system with an outer giant population. The classification matters because it gives future surveys and formation models a concrete, observationally grounded set of target architectures to explain.","feed_headline":"Three questions classify 97% of multiplanet systems","feed_subtitle":"A new scheme sorts the exoplanet zoo into peas-in-a-pod, warm-Jupiter, and gapped systems, with the Solar System in the majority.","key_machinery":"The machinery is a three-question classification tree applied to the 314 confirmed systems with at least three planets, built on radius classes (Earths, sub-Neptunes, Neptunes, Jupiters) and period-ratio thresholds. The load-bearing definitions are: an inner/outer divide requires a period ratio $>5$ between adjacent planets with the outer planet a Jupiter beyond 130 days; a “gapped” system has any inner pair with period ratio $>5$; “peas-in-a-pod” means no inner Jupiters. The tree then splits gapped systems by whether the gap lies between the innermost pair, the middle, or the outermost pair, and this location distinction is argued to be physically meaningful because inner gaps track ultra-short-period planets while outer gaps track cooler exterior planets.","core_discovery":"The central claim is that the architecture of a planetary system can be captured by a short decision tree. First, does the system have distinct inner and outer regimes, defined by a period gap with ratio $>5$ where the outer planet has period $>130$ days and is Jupiter-sized? Second, among the inner planets, is there at least one Jupiter (radius $>6\\,R_\\oplus$)? Third, do the inner planets contain a gap with period ratio $>5$? Answering these questions assigns roughly 97% of confirmed $N \\geq 3$ systems to a category: closely-spaced peas-in-a-pod, gapped peas-in-a-pod (inner-, middle-, or outer-gap), or warm Jupiter systems, with hot Jupiters and strongly-inverted mass ratios as auxiliary dynamical features. The same lens makes the Solar System typical: an inner peas-in-a-pod group of small planets, separated by a wide gap from outer giants. The paper presents this as a working classification for the current catalog, explicitly qualitative in its statistics, and argues that the existence of the categories is robust even if individual assignments can shift as data improve.","pith_inferences":["If the architecture classes are real, population synthesis models should be constrained to reproduce the observed 80% closely-spaced peas-in-a-pod fraction and the roughly 39% gap rate among warm Jupiter systems, not just broad occurrence rates.","The 130-day and period-ratio-5 boundaries may track the ice line and giant-planet migration; one testable extension is whether gap location shifts with stellar mass as the ice line moves, which the paper's period-only definition does not capture.","The framework predicts that upcoming microlensing surveys will find non-Jovian outer planets, like Uranus and Neptune analogs, filling the paper's one conspicuously empty class; their absence would instead suggest a real formation barrier.","If the peas-in-a-pod pattern extends to M-dwarfs, their compact chains can place planets in the habitable zone, so the framework focuses future habitability searches on M-dwarf multis; this follows from the paper's claims but goes beyond its classification result."],"forward_implications":["Roughly 80% of systems with at least three inner planets are closely-spaced peas-in-a-pod systems, making that the default outcome of planet formation.","Warm Jupiter systems are a minority (about 8% of $N \\geq 3$ inner systems) but are much more likely to be gapped and to show strongly inverted mass ratios, pointing to different formation histories.","Large gaps among inner planets usually sit at the inner or outer edge of the system, not the middle, so inner-gap and outer-gap systems are separate dynamical subclasses.","Hot Jupiters rarely have close companions, but about 3% have distant cold Jupiter companions, consistent with high-eccentricity migration as a dominant formation channel.","The Solar System, with its small inner planets and outer giants, fits the standard peas-in-a-pod-plus-outer-Jupiters pattern rather than being an oddity."],"supporting_citations":[{"why":"Supplies the complete confirmed-planet catalog, the Planetary Systems Composite Parameters Table, that the classification is built on.","marker":"NASA Exoplanet Science Institute 2020"},{"why":"Provides the mass-radius relation used to fill missing radii and masses and to recalculate unphysical TTV-derived values during data filtering.","marker":"Chen & Kipping 2017"},{"why":"Defined the peas-in-a-pod pattern of similar-sized, regularly spaced planets that the paper adopts as its main category.","marker":"Weiss et al. 2018"},{"why":"Supplies the compact multiplanet system concept and the comparison population for the paper's closely-spaced peas-in-a-pod class.","marker":"Weiss et al. 2023"},{"why":"Prior information-theoretic architecture classification whose gap-complexity split is compared and contrasted with the paper's gap-based categories.","marker":"Gilbert & Fabrycky 2020"},{"why":"Prior a priori architecture classes (ordered, anti-ordered, mixed, similar) that this framework refines by adding period spacing.","marker":"Mishra et al. 2023a"},{"why":"Numerical stability criteria for planets in binaries provide the working estimate that motivates the period-ratio threshold near 5 for dynamical detachment.","marker":"Holman & Wiegert 1999"},{"why":"Statistical evidence for an outer edge of closely-packed peas-in-a-pod systems, supporting the paper's assumption that observed gaps are largely real.","marker":"Millholland et al. 2022"},{"why":"Supplies the high-eccentricity migration picture used to interpret hot Jupiter companions and the 3% cold-Jupiter companion rate.","marker":"Zink & Howard 2023"},{"why":"Establishes the radius valley that motivates the paper's four planet-size classes.","marker":"Fulton et al. 2017"}],"fun_headline_variants":["Exoplanet architectures sorted into a simple decision tree","Nearly all multiplanet systems fit three-question scheme","Peas-in-a-pod and warm Jupiters: a new exoplanet taxonomy","97% of multiplanet systems classed by three questions","New framework maps the exoplanet zoo"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The classification only works as a statement about nature if the chosen thresholds—a period ratio above 5 for a gap and 130 days for an outer planet—carve at physical joints rather than being boundaries fitted to today's catalog; the paper itself concedes this definition is “both arbitrary and over-fit to our small sample.”","fun_headline_variants_meta":{"raw":{"variants":["Exoplanet architectures sorted into a simple decision tree","Nearly all multiplanet systems fit three-question scheme","Peas-in-a-pod and warm Jupiters: a new exoplanet taxonomy","97% of multiplanet systems classed by three questions","New framework maps the exoplanet zoo"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00021,"raw_usage":{"total_tokens":1437,"prompt_tokens":998,"completion_tokens":439,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":614,"completion_tokens_details":{"reasoning_tokens":354}},"tokens_in":614,"tokens_out":439,"duration_ms":4164,"temperature":1.0,"reasoning_tokens":354,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T20:29:06.386882+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Re-run the classification on the same 314 systems with the inner/outer cutoff moved to 80 days and 200 days and the gap threshold moved to 4 and 6; the framework's claim would fail if many systems change class or if the fraction classified “with minimal ambiguity” drops well below the reported 97%. A second decisive test is the discovery of a non-Jovian outer planet, which the framework predicts should be rare enough to leave a conspicuous empty class.","supporting_citations":[],"review_version":1}