{"id":"c78a1399-32fb-48fb-b0f1-ec2fb12f27d0","arxiv_id":"1908.07342","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"New flat-back 3D gadgets for origami extrusions are constructed and shown to be downward compatible with pyramid-supported gadgets, enabling prism extrusions at least 4/3 times taller, and sqrt(2) times for triangles.","lead":"This paper introduces a new class of 3D origami gadgets that create raised faces on a flat sheet, replacing older pyramid-supported designs. The new gadgets have flat backs, allow taller extrusions, and can often be swapped into existing crease patterns without changing the outgoing pleats.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Height-ratio theorem depends on an unproved global consistency of the per-edge minima in Theorem 5.7; local interference coefficients alone do not establish h_new.","rationale":"The reader's weakest assumption correctly identifies the unproved global choice of the order of outgoing pleats in Theorem 5.7. My stress-test pass converges on that same point as the most load-bearing gap for the central quantitative claim. The 4/3 and sqrt(2) height ratios are the strongest advertised results, and they depend on h_new being exactly the edge-wise minimum formula. The paper gives careful local angle and length computations, and the interference-coefficient inequalities of Theorem 7.3 are plausible and independently checkable, but they only bound the ratio of coefficients. They do not prove that the minimizing choices for every edge can be realized by a single global folding order. If the choices are per-gadget rather than per-edge, an odd polygon such as a triangle can create a parity obstruction; if they are per-edge, an explicit layer-ordering construction is missing. The division/repetition constructions in Section 8 are explicitly stated without proof, but they are auxiliary to the headline height-ratio theorem, so I do not treat them as the primary concern. My recommendation is to keep the reader's CONDITIONAL verdict: the paper is promising and largely self-consistent, but the headline efficiency gain requires an additional global realizability argument or a counterexample. Hence UNCHANGED.","tokens_in":33743,"tokens_out":6384,"duration_ms":69614,"concrete_test":"Take a triangular prism with all alpha_i = pi/2, the limiting case of Theorem 7.3. Enumerate all possible global assignments of inner/outer status to the six outgoing pleats of the three gadgets. For each assignment, compute the minimal clearance along each bottom edge using the length formulas in Proposition 5.2 and Definition 5.5; take the maximum over assignments. Compare this true h_new with sqrt(2) h_conv, and repeat the same enumeration for a square prism to compare with 4/3 h_conv. If no assignment attains min_i |B_iB_{i+1}| / kappa_new(B_iB_{i+1}), then Theorem 5.7's formula overestimates h_new and Corollary 7.5 is not established; if a consistent assignment does attain the claimed value for both the triangle and the square, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim (Corollary 7.5) is h_new > 4/3 h_conv for any convex polygon, and h_new > sqrt(2) h_conv for any triangle. This is obtained by combining Theorem 5.7 with the interference-coefficient bounds in Theorem 7.3. Theorem 5.7 asserts h_new = min_i |B_iB_{i+1}| / kappa_new(B_iB_{i+1}) only 'if we choose the order of the outgoing pleats appropriately.' Definition 5.5 defines kappa_new(B_iB_{i+1}) as a minimum over two possible local orders. Taking that minimum edge-by-edge does not prove that one global folding order realizes all the minimizing choices simultaneously. If the inner/outer status of a pleat is a property of a whole gadget, the choices on the two sides of a gadget are coupled, and around an odd cycle such as a triangle the per-edge minima can become mutually inconsistent; if it is a per-edge property, the paper supplies no construction of a layer ordering that realizes it. In either case the formula for h_new is an upper bound on the true maximum height, not a proven value. Corollary 7.5 inherits this gap: the 4/3 and sqrt(2) ratios compare the true conventional maximum with this unproved new maximum. This is load-bearing because a shortfall in h_new from inconsistent choices could push the ratio below 4/3 in exactly the limiting cases where Theorem 7.3's supremum is approached. Local flat-foldability checks and the coefficient inequalities in Theorem 7.3 do not resolve this, because they address adjacent gadgets pairwise rather than a simultaneous global layer ordering.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a new family of 3D gadgets for origami extrusions, characterized by simple outgoing pleats, flat back sides above the ambient paper, and downward compatibility with the conventional pyramid-supported gadgets of Natan. The central quantitative claim is that for a prism over any convex polygon the maximal extrudable height with the new gadgets is more than 4/3 times that with the conventional gadgets, and more than sqrt(2) times for triangular prisms. The construction algorithms are explicit, and the analysis is based on newly defined interference coefficients that are derived in closed form. The paper also presents, without proof, constructions for division/repetition of the new gadgets and sketches negative gadgets and curved-crease variants.","tokens_in":34017,"tokens_out":7044,"duration_ms":77981,"significance":"The paper addresses a concrete and active problem in computational origami, and if its main height-ratio theorem is correct, it gives a clean quantitative design rule that improves on an established method. The strengths are real: the constructions are algorithmic and reproducible, the interference-coefficient formulas are explicit and parameter-free, and the comparison with conventional gadgets uses a fixed external baseline rather than fitted data. The flat-back property and the claimed compatibility are interesting in their own right. However, the central claim rests on a global consistency of per-edge choices in the height-maximization theorem, and that consistency is not proved. Because the same gap propagates to the headline 4/3 and sqrt(2) bounds, the paper cannot currently be accepted as a proof of its main quantitative result.","major_comments":[{"comment":"The formula for h_new in Theorem 5.7 is not established. Definition 5.5 defines kappa_new(BiBi+1) as the minimum over two local choices of which adjacent pleat is inner and which is outer, but the theorem's caveat 'if we choose the order of the outgoing pleats appropriately' is not proved: a single global folding order (or any consistent layer assignment) must realize the minimizing choice on every edge simultaneously. For an odd cycle such as a triangle, the per-edge minima can correspond to a cyclic orientation that no total order of the gadgets realizes. In that case the right-hand side of Theorem 5.7 is only an upper bound on the true maximum height, not the claimed value. Since Corollary 7.5 compares this quantity with the conventional maximum, the 4/3 and sqrt(2) ratios inherit the gap. A concrete test would be to determine, for a nearly equilateral triangle, whether a single assignment attains all three edge-wise minima; if not, the claimed improvement may be overstated.","section":"Theorem 5.7 / Definition 5.5"},{"comment":"The foldability argument for the new crease pattern is local only. The paper checks Kawasaki angle sums around vertices of the lower part and asserts that the upper part folds as the intended side and top faces, but it does not prove that the full assembled crease pattern folds into the intended 3D extrusion without self-intersection. The interference coefficients in Section 5 detect collisions between adjacent gadgets along shared bottom edges, but they do not rule out other self-intersections of the assembled pattern. Thus the existence of the extruded polyhedron at the claimed maximal height is not rigorously established, and this is load-bearing for Corollary 7.5 as well.","section":"Section 4 / Section 5"}],"minor_comments":[{"comment":"Construction 8.1 and the accompanying mountain/valley tables are presented 'without proof.' Since the abstract lists division/repetition gadgets as a contribution, the claims should be flagged as conjectural or supported by a foldability proof or simulation.","section":"Section 8"},{"comment":"There are numerous typographical errors, including 'Calros Natan' in the abstract, 'delolopment', 'appropiately', 'spremum', 'fuctions', and 'efieient'. These do not affect the mathematics but should be corrected.","section":"Throughout"},{"comment":"References [6] and [8] point to personal Flickr albums rather than stable archival sources. This is particularly problematic for [8], which is used as the baseline for the conventional gadgets; a more stable reference or a formal description would improve verifiability.","section":"References [6], [8]"},{"comment":"Remark 5.9 suggests 'folding back' or 'sinking' to avoid a certain interference, but this is not formalized. Since it is offered as a solution to a possible failure mode, a precise description of the added creases and a proof of their validity would be useful.","section":"Remark 5.9"}],"recommendation":"major_revision","confidential_remarks":"The central quantitative claim is not yet proved because of the unresolved global consistency of the per-edge minima in Theorem 5.7. If the author can supply a proof that a single global order attains all edge-wise minima (or a counterexample showing the formula is only an upper bound), the paper would be substantially stronger. I would also encourage a simulation or explicit folded example for a triangular prism, since that is the case where the odd-cycle difficulty is most acute."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: Doi has a genuinely new family of 3D gadgets for origami extrusions, with a clear construction, an interference-coefficient framework, and proved height-ratio bounds (4/3 for convex prisms, sqrt(2) for triangular prisms) against Natan's pyramid-supported gadgets. The paper is worth refereeing. The main weakness is that global foldability of the new crease patterns is asserted rather than proved: the checks are local (Kawasaki) plus pairwise interference coefficients, and Theorem 5.7 depends on a phrase \"if we choose the order of the outgoing pleats appropriately\" that never gets a formal proof.\n\nWhat's new: the flat-back design, downward compatibility (Theorem 6.2), the interference coefficients (Definitions 5.4/5.5), and the comparison theorems (7.1–7.3, Corollary 7.5). The geometry is worked out carefully, and the length calculations in Propositions 5.2 and 5.3 check out. The proof of Theorem 7.3 is intricate but the averaging argument for the supremum is valid. The incircle formula for triangles (Theorem 7.1) is elegant.\n\nSoft spots, in order of importance. First, the paper never proves that a crease pattern satisfying the local flat-foldability and per-edge interference conditions folds into the intended 3D shape without self-intersection. That's a real gap, though not unusual in this literature. Second, Section 8 (division/repetition) is explicitly presented \"without proof\"; it's clearly labeled, so not a hidden flaw, but it limits completeness. Third, the flat-foldable cube and curved-crease examples in Section 9 are illustrative. The stress-test worry that per-edge minima might be mutually inconsistent around an odd cycle does not, on reading, hold up: each side of a gadget appears in exactly one edge coefficient, so the choices are independent per side and the min can be taken edge-by-edge. The real issue is that the paper gives no explicit layer ordering realizing those choices—i.e., no global foldability argument.\n\nWho it's for: people working on origami extrusions, tessellations, and computational origami design. It's a solid specialized contribution with honest (mostly local) proofs and one clear unproved step. A serious referee should engage; the paper likely needs a revision that either proves global foldability under stated conditions or explicitly reframes the height-ratio theorems as bounds conditioned on foldability. I'd accept it for review.","headline":"New flat-back origami extrusion gadgets with clean height-ratio theorems; the main gap is global foldability, argued only locally.","tokens_in":34572,"tokens_out":4900,"would_cite":false,"duration_ms":50945,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper introduces flat-back 3D gadgets for origami extrusions that replace pyramid-supported gadgets and push the maximal extrudable prism height past 4/3 of the old bound.","keywords":["origami extrusion","3D gadget","simple pleats","flat-foldable","interference coefficient","prism height","crease pattern"],"falsifier":"Fold or rigid-fold-simulate the crease pattern of a 1x1 square prism extruded to height sqrt(2) with the new cube gadgets shown in Figure 1.3; if the panels self-intersect, the pattern jams, or the back sides are not flat, then the claimed height bound and the downward-compatibility theorem fail for that configuration.","tokens_in":33499,"feed_emoji":"📐","tokens_out":8123,"duration_ms":81943,"temperature":0.7,"pith_summary":"An origami extrusion raises a solid shape out of the middle of a flat sheet, and the local crease patterns that do this are called 3D gadgets. This paper introduces a new family of such gadgets whose back sides are flat above the paper, replacing the usual pyramid-supported gadgets that carry an internal supporting pyramid. The new gadgets are downward compatible: in most cases they can be swapped into an existing crease pattern with the same outgoing pleats, and they never need more room along the shared bottom edges because their 'ears' and 'tongue' occupy less space. The paper proves that the maximal height of a single-step extrusion of a prism over any convex polygon is more than 4/3 times the height reachable with the conventional gadgets, and for triangular prisms more than the square root of two times that height. This matters because height in an origami extrusion is limited by interference between adjacent gadgets, and the new construction relaxes exactly that bottleneck.","feed_headline":"New origami gadgets extrude prisms a third higher","feed_subtitle":"Flat-back gadgets replace pyramid supports, so taller prisms can be folded in a single extrusion.","key_machinery":"The workhorse is the interference coefficient, a normalized length that measures how much room a gadget needs along a shared bottom edge. For the conventional gadgets it is the total length taken up by the two internal supporting pyramids at unit height, while for the new gadgets it is the minimum of two sums, one combining the inner-pleat coefficient of one gadget with the outer-pleat coefficient of the other and the other combining them in the opposite order. The maximal height of an extrusion is the minimum over bottom edges of edge length divided by the relevant interference coefficient, so every comparison of heights reduces to comparing these coefficients. The geometry that makes the new coefficients small is the ears-and-tongue construction: two kites fold flat over the side faces, replacing the bulky triangular pyramid, and the smaller footprint leaves more room for adjacent gadgets. This machinery also carries the downward-compatibility proof, because with the pleat-angle adjustments set to zero, each new coefficient is at most the corresponding conventional coefficient.","core_discovery":"The central discovery is a construction (Construction 3.2) of 3D gadgets with simple outgoing pleats, each a parallel mountain-valley pair, that unlike the conventional pyramid-supported gadgets have flat back sides and allow the two outgoing pleat angles to be adjusted independently by small amounts. The construction replaces the internal supporting pyramid with two 'ears' and a 'tongue' that fold flat against the side faces, giving interference coefficients that are pointwise at most those of the conventional gadgets when the pleat angles are unchanged. Because a gadget fits along a bottom edge exactly when height times the relevant interference coefficient is at most the edge length, smaller coefficients translate directly into taller extrusions. For prisms over any convex polygon the paper computes the ratio of new to conventional interference coefficients to be less than 3/4, with supremum 3/4, and for triangles less than 1/sqrt(2), with supremum 1/sqrt(2), yielding maximal heights greater than 4/3 and greater than sqrt(2) times the conventional maxima, respectively. The same construction also yields flat-foldable extrusions, negative gadgets, and a division and repetition scheme for stacking gadgets to gain height.","pith_inferences":["The paper's verification is local, so a natural next step is a rigid-origami simulation of the Figure 1.3 pattern to test whether the claimed sqrt(2)-height square prism folds without global self-intersection.","Because the 4/3 and sqrt(2) factors are suprema approached in limiting angle configurations, practical gains will vary with polygon shape, and designers could choose polygon angles to sit near the high-ratio regime.","The negative-gadget constructions suggest a route to extruding shapes with valleys or reentrant solid angles, such as the regular octahedron and icosahedron mentioned in the conclusion, if the local flat-foldability can be upgraded to a global folding certificate.","The division and repetition section implies a divide-and-conquer design strategy: split a tall extrusion into lower gadgets to shrink the interference footprint per layer, and testing the stacking rule on a square prism taller than 1x1x5 would show whether the claimed lack of upper-gadget interference holds in a physical fold."],"forward_implications":["Any crease pattern built from conventional pyramid-supported gadgets can in most cases be rebuilt with the new gadgets using the same outgoing pleats, and the extrusion height never has to decrease; the only excluded case is when one of the two inequalities beta_L + gamma/4 < pi/2 or beta_R + gamma/4 < pi/2 holds.","For a prism over any convex polygon, a single application of the new gadgets extrudes more than 4/3 times the maximal conventional height, and for triangular prisms the factor exceeds sqrt(2).","Flat back sides make it possible to add twist creases for flat-foldable extrusions and to deform extrusions with curved creases, which the conventional pyramid support prevents.","The independent pleat-angle parameters delta_L and delta_R give designers freedom to route outgoing pleats around neighbouring extrusions, subject to the conditions delta_L, delta_R > 0 and delta_L + delta_R < pi - gamma.","Repetition and proportional division of the new gadgets can stack layers to reach heights beyond the one-step bound while keeping interference distances fixed, although the division constructions are presented without proof."],"supporting_citations":[{"why":"Identifies the cube gadget that the new cube gadget replaces in the opening example and supplies the polycube extrusion setting.","marker":"[1]"},{"why":"Provides the alternating angle-sum criterion for local flat-foldability used to verify each vertex of the new crease patterns.","marker":"[7]"},{"why":"Defines the conventional pyramid-supported 3D gadgets that the new gadgets are compared with, replaced, and shown downward compatible with.","marker":"[8]"}],"fun_headline_variants":["Flat-back origami gadgets raise prism heights by a third","New flat-back folds make origami prisms 4/3 taller","Origami gadgets with flat backs boost extrusion height","Replace pyramid supports: origami prisms grow 33%","Flat-back 3D gadgets yield taller origami prisms"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The construction rests on the assumption that local flat-foldability around every vertex, together with the interference-coefficient inequalities, guarantees the whole crease pattern folds into the intended 3D extrusion without self-intersection.","fun_headline_variants_meta":{"raw":{"variants":["Flat-back origami gadgets raise prism heights by a third","New flat-back folds make origami prisms 4/3 taller","Origami gadgets with flat backs boost extrusion height","Replace pyramid supports: origami prisms grow 33%","Flat-back 3D gadgets yield taller origami prisms"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000241,"raw_usage":{"total_tokens":1599,"prompt_tokens":1099,"completion_tokens":500,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":715,"completion_tokens_details":{"reasoning_tokens":416}},"tokens_in":715,"tokens_out":500,"duration_ms":5750,"temperature":1.0,"reasoning_tokens":416,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:19:46.773329+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Fold or rigid-fold-simulate the crease pattern of a 1x1 square prism extruded to height sqrt(2) with the new cube gadgets shown in Figure 1.3; if the panels self-intersect, the pattern jams, or the back sides are not flat, then the claimed height bound and the downward-compatibility theorem fail for that configuration.","supporting_citations":[{"cited_title":"Benbernou, Erik D","cited_arxiv_id":null,"evidence_quote":"Identifies the cube gadget that the new cube gadget replaces in the opening example and supplies the polycube extrusion setting."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the alternating angle-sum criterion for local flat-foldability used to verify each vertex of the new crease patterns."},{"cited_title":"11-9-302 Y UMOTO -CHO , TAKARAZUKA , H YOGO 665-0003, J APAN E-mail address: doi.mamoru@gmail.com","cited_arxiv_id":null,"evidence_quote":"Defines the conventional pyramid-supported 3D gadgets that the new gadgets are compared with, replaced, and shown downward compatible with."}],"review_version":1}