---
title: 'COrigami: AI Pipeline & Curved-Crease Mechanics'
url: https://www.emergentmind.com/topics/corigami
type: topic
---

# COrigami: AI Pipeline & Curved-Crease Mechanics

Searching arXiv for the named COrigami paper and closely related curved-crease origami work.
Search query: COrigami origami AI pipeline flat-foldable visually recognizable origami arXiv
COrigami denotes two related but distinct usages in origami research. In the contemporary computational-design literature, it refers specifically to **“COrigami: An AI Pipeline for Co-Designing Flat-Foldable Visually Recognisable Origami”**, a neuro-symbolic, end-to-end system that generates flat-foldable, visually recognizable origami from natural-language prompts [2606.19177]. In mechanics and geometry, “COrigami” is also used as shorthand for **curved origami** or **curved-crease origami**, where curvature emerges either from explicitly curved fold lines or from elastic incompatibility in straight crease networks [1206.0461; 2102.03753; 2412.18265]. The former usage concerns AI-assisted co-design under flat-foldability constraints; the latter concerns the mechanics, kinematics, and morphogenesis of folded curved structures. The term therefore spans both a recent AI pipeline and an established research area in origami mechanics, linked by a common emphasis on mathematically constrained shape generation.

## 1. Terminological scope and research contexts

The AI-system usage of COrigami is explicit and title-level: the 2026 paper presents “an end-to-end AI-driven pipeline that assists the design cycle by generating crease patterns from natural language” [2606.19177]. Its stated domain is **computational origami**, described as “a mathematically rigid environment that grounds artistic design within the equations of flat foldability” [2606.19177]. In this setting, COrigami is a co-design framework rather than a geometric subfield.

A separate usage appears in the mechanics literature, where the 2021 paper on accordion-like elastic folding states that one can generate **“curved origami, or COrigami,”** without explicitly drawing a curved crease [2102.03753]. There, COrigami means a curved deployed shell generated by a straight crease network, with curvature emerging from **mechanical frustration** rather than from intrinsic geodesic curvature in the crease pattern [2102.03753]. Closely related work on **curved crease origami** analyzes folding along curved lines as a geometric-mechanical problem governed by developability, crease stiffness, and geometric frustration [1206.0461; 2412.18265].

These usages are not identical. The AI pipeline is concerned with text-conditioned generation of **flat-foldable** and **visually recognizable** designs [2606.19177]. The mechanics usage concerns curved folds, distributed actuation, and shell morphogenesis [1206.0461; 2102.03753; 2412.18265]. A plausible implication is that the shared term reflects a broader view of origami as a domain where geometry, mechanics, and algorithmic synthesis are tightly coupled.

## 2. COrigami as an AI pipeline for flat-foldable recognizable origami

The 2026 COrigami system is described as a **neuro-symbolic, end-to-end AI system** for co-designing **flat-foldable, visually recognizable origami** from natural-language prompts [2606.19177]. Its motivating claim is that origami is an unusually difficult testbed for generative AI because it combines **strict geometric feasibility** with **subjective aesthetic/semantic fidelity** [2606.19177]. The paper argues that these goals are often in tension and therefore should not be handled by unconstrained end-to-end generation.

The pipeline is explicitly staged as follows: **natural-language prompt → semantic stick figure → discrete base packing on a box-pleated grid → flat-foldable crease pattern → shaped 3D origami → autonomous VLM aesthetic evaluation → RL refinement of shaping decisions using that evaluation as reward** [2606.19177]. The architecture divides labor between language- and vision-language-model components for semantic conceptualization and aesthetic critique, and deterministic geometric algorithms for packing, tiling, foldability checking, and folding simulation [2606.19177].

The paper states that the system acts as “a highly effective collaborative assistant, generating structural starting points that human artists can further expand and shape” [2606.19177]. It also states that human experts remain essential for final physical realization because the simulated designs ignore **paper thickness, bulking, and craft-level post-processing** [2606.19177]. This places COrigami within co-creative design rather than autonomous fabrication.

The system is motivated in part by a negative result: direct fine-tuning of a language model for raw SVG crease-pattern generation “saturates at only about **60% flat foldability**” [2606.19177]. The staged pipeline is therefore presented not as a stylistic preference but as a response to brittleness, long-output consistency, scarce data, and the difficulty of automatically defining aesthetic quality [2606.19177].

## 3. Internal pipeline: semantic abstraction, packing, solving, and shaping

The first stage converts text prompts into a **semantic stick figure**, defined as a tree-structured, acyclic skeleton encoding the intended object in terms of labeled nodes and edges [2606.19177]. Each edge is parameterized by **a unique label, length, azimuth angle, and elevation angle** [2606.19177]. Leaf nodes correspond to **flaps**, and internal edges correspond to **rivers** [2606.19177]. The system uses constrained prompting to enforce properties such as **symmetry**, **no graph cycles**, and **anatomically plausible part structure**, then uses Gemini as a VLM to inspect rendered views from **top, side, front, and isometric** angles [2606.19177].

The second stage maps the stick figure to a square sheet through a **discrete rectangle-packing and tiling problem** on an integer grid [2606.19177]. COrigami adopts a **box-pleated** representation in which axis-parallel creases and hinges lie on an orthogonal integer grid and diagonal ridges are limited to \(45^\circ\) [2606.19177]. Leaf nodes become rectangles, internal edges become proportional-width paths, and rivers partition the paper into **pockets** [2606.19177]. The packing solver performs iterative backtracking over **river placement, flap placement, pocket filling, symmetry constraints, overlap checks, and area-feasibility checks**, with the first river placed by exhaustive enumeration and later rivers using a wall-following strategy [2606.19177]. After a feasible layout is found, a tiling step expands adjacent flaps to eliminate gaps so that the result is a **complete contiguous tiling** [2606.19177].

The third stage deterministically converts the packing into a flat-foldable crease pattern [2606.19177]. The paper distinguishes **local flat foldability** from **global flat foldability**. For local flat foldability it explicitly invokes **Kawasaki’s theorem**, requiring alternating sums of sector angles to equal \(180^\circ\), and **Maekawa’s theorem**, requiring \(|M-V|=2\) at an interior flat-foldable vertex [2606.19177]. It also uses a recursive **crimping algorithm** as a sufficient local test [2606.19177]. For global flat foldability, the paper states that the system uses the **facewise formulation** from FlatFolder, treating the problem as a finite constraint-satisfaction graph over overlapping convex faces, with propagation through precomputed tables and DFS backtracking on remaining connected components [2606.19177].

The deterministic solving stage is described as a sequence of **pleat construction, pleat interleaving assignment, ridge construction and assignment, hinge assignment, and reassignment of trapped pleats as needed** [2606.19177]. Pleats are assigned alternating mountain/valley orientations by BFS over connected paths, while diagonal \(45^\circ\) ridges are propagated from anchor points such as Y-shaped vertices and boundary conditions [2606.19177]. Hinge assignment is treated as a combinatorial search over standard interleaved \(M\!-\!V\!-\!M\!-\!V\) and symmetric \(M\!-\!V\!-\!V\!-\!M\) orientations [2606.19177].

The shaping stage occurs after flat-foldable synthesis. First, **tree shaping via simple folds** reconstructs the 3D skeleton of the stick figure [2606.19177]. The paper defines a simple fold by a cut line through two points \(p_1,p_2\), assigned mountain or valley orientation [2606.19177]. The shaping algorithm traverses the tree in BFS order, computes each child orientation relative to its parent, and applies a physically realizable fold axis lying in the parent paper plane [2606.19177]. Second, a **clip pattern algorithm** handles **narrowing**, propagating a local 2D reference frame across layers, projecting shaping templates onto folded layers by affine transformation, detecting \(Z\)-axis flips, and clipping all lines to the hull of each face [2606.19177]. The paper distinguishes **symmetric narrowing** from **asymmetric narrowing** and notes that short rivers may be too short to narrow with this method [2606.19177].

## 4. Reinforcement learning and autonomous aesthetic evaluation

A distinctive feature of COrigami is its reinforcement-learning stage guided by a vision-language evaluator [2606.19177]. The paper states that the heuristic tree-shaping stage reproduces the skeleton faithfully but not necessarily the best-looking or most semantically faithful final object, so RL is used to explore a broader shaping space [2606.19177].

The policy model is **Gemini 2.5 Flash Lite**, and it outputs tool-use parameters for all flaps in one step rather than sequentially [2606.19177]. The action space includes **simple folds, narrowing, and additional shaping variations** [2606.19177]. The reward combines a hard invalidity penalty, VLM-derived feedback, and a small intrinsic reward for action diversity [2606.19177]. The paper gives the intrinsic term as
\[
r_i = \min\left(\frac{n}{10}, 1\right) * 0.6
\]
where \(n\) is the number of successful tool calls [2606.19177]. Invalid trajectories or failures receive \(r=-1\) [2606.19177]. Training uses **batch size \(64\)**, **learning rate \(10^{-4}\)**, a **simple policy-gradient algorithm**, **KL regularization to the base policy**, and a KL coefficient decayed from **1 to \(10^{-4}\) over 500 steps** [2606.19177].

The evaluator is **Gemini 3 Flash** with **temperature 0** and no majority voting in the default deployment [2606.19177]. In **Single Model Evaluation**, the evaluator receives the prompt and **seven rendered views** of one candidate, then scores **appendage count, topology, proportionality, differentiation between body regions, and aesthetic refinement** on a \(0\) to \(10\) scale later normalized to \([0,1]\) [2606.19177]. In **Comparison Judge Mode**, it compares reference and candidate images and chooses the better representation, with swapped ordering to reduce proximity bias [2606.19177]. The best-performing deployment for curation is described as a **“double tournament”** style comparison [2606.19177].

This coupling of deterministic validity checking with autonomous aesthetic critique is central to the paper’s claim that AI can support mathematically constrained artistic co-design [2606.19177].

## 5. Experimental results, benchmarks, and limitations

The paper benchmarks direct end-to-end generation against the staged pipeline. For the direct SVG-space baseline, a Gemini model is trained on **400k synthetic crease patterns** totaling approximately **3.2B tokens** [2606.19177]. The reported outcome is that syntax validity and flat foldability improve during early training, but test-set flat foldability plateaus at **about 60%** [2606.19177].

For the VLM evaluator benchmark, the paper uses **87 positive examples** and **152 negative examples** [2606.19177]. It reports that **Gemini Flash** outperforms Gemini Pro, that prompt engineering matters substantially, and that the **Double** tournament setup is best, achieving **0.811 accuracy**, **0.651 AP**, and **0.74 \(F_1\)** [2606.19177].

For the full pipeline, the paper reports that from **560,000 initial tree candidates**, COrigami yields **113,276** valid semantic stick figures [2606.19177]. It then reports a **packing pass rate of 55.3%**, **solving pass rate of 79.2%**, **shaping pass rate of 92.0%**, **27,869 final curated baseline models**, and an **overall survival rate of 5.0%** [2606.19177]. It further states that **17,789 designs** were filtered out during final verification, including **7,490** due to low VLM reward and **10,299** due to failing a tree-similarity threshold [2606.19177]. Failure is reported to increase with **more flaps**, **more rivers**, and **denser trees**, with packing and solving identified as the main bottlenecks for complex topologies [2606.19177].

The deterministic folding engine is compared with a GPU mass-spring simulator on **87 complex crease patterns** [2606.19177]. The paper states that the deterministic method achieves significantly lower reconstruction error, “in some cases by up to **five orders of magnitude**,” with vertex errors as low as \(10^{-5}\) compared to a baseline around \(10^{-1}\) [2606.19177].

The limitations are stated explicitly. The current shaping repertoire mainly includes **simple folds** and **narrowing**, while more advanced methods such as **Pythagorean stretches** and **level shifters** are not integrated [2606.19177]. The simulator is geometric rather than thickness-aware and does not model **layer bulk, fiber compression, paper creep,** or **tearing limits** [2606.19177]. Dataset scarcity is also emphasized, including a foundational dataset of about **100 visually recognizable traditional origami models** created with collaborating designers [2606.19177]. The paper also notes that the final top-10 figure involved human selection from RL samples by visual inspection of high-reward outputs [2606.19177]. These statements delimit the present system as a structural blueprint generator rather than a complete replacement for expert folding practice.

## 6. COrigami as curved origami and curved-crease mechanics

In the mechanics literature, COrigami refers to the generation or analysis of curved folded structures. The 2021 paper on an accordion-like fold network states that its contribution is to show that one can generate **curved origami**, or **COrigami**, without explicitly drawing a curved crease [2102.03753]. The studied system consists of a thin elastic sheet with **one central longitudinal mountain crease**, a sequence of **\(N\) equally spaced transverse creases** crossing it perpendicularly, alternating mountain/valley directions, and rectangular facets of dimensions \(l\times w\) [2102.03753]. Although the imprinted crease network contains no geodesic curvature, the deployed structure exhibits an effective curvature generated by the deformed central fold [2102.03753].

The paper defines a discrete curvature
\[
\kappa = \frac{1}{R} = \frac{\sum_{i=1}^N \delta\theta_i}{(N+1)w},
\]
where the \(\delta\theta_i\) are angular mismatches between successive central-crease segments [2102.03753]. Curvature is attributed to **competition between crease stiffness and the kinematics of the crease network**, producing **mechanical frustration** [2102.03753]. It reports the scaling laws
\[
R_\text{simu} \propto w^{2/3}l^{1/3}, \qquad R_\text{exp} \propto w
\]
for simulations and experiments respectively [2102.03753]. It also identifies three deformation regimes organized primarily by the aspect ratio \(l/w\): **Region I: single-facet deformation** for \(l<l_1\approx 1.5w\), **Region II: faceting** for \(l_1<l<l_2\), and **Region III: buckling** for \(l>l_2\approx 9w\) [2102.03753]. The corresponding morphogenesis progresses from localized ridge-like bending to triangular faceting and then Euler-like edge buckling [2102.03753].

A more classical curved-crease formulation appears in the theory of an annular elastic strip folded along a central circular crease [1206.0461]. That paper treats curved crease origami as a mechanics problem for a thin annulus of thickness \(t\), width \(2w\), and crease radius \(r\), in the regime
\[
t \ll w < 2\pi r
\]
[1206.0461]. The crease becomes a space curve with curvature \(\kappa(s)\), torsion \(\tau(s)\), and dihedral angle \(\theta(s)\), under the assumption that the sheet is **isometrically deformed everywhere except at the crease** [1206.0461]. A central geometric relation is
\[
\sin\!\left(\frac{\theta}{2}\right)=\frac{1}{\kappa},
\]
with \(\kappa\) scaled by \(r\) [1206.0461]. For an actual fold with \(\theta<\pi\), one obtains \(\kappa>1\), leading to the conclusion that a **closed** circular crease cannot remain planar and therefore **buckles out of plane** [1206.0461]. The total energy is written as
\[
E = E_b + E_c,
\]
with panel bending energy \(E_b\) and crease energy \(E_c\), the latter penalizing deviations from a preferred dihedral angle \(\theta_0\) via the parameter \(\sigma = Kr/B\) [1206.0461]. The paper distinguishes **stiff folds**, for which narrow stiff creases exhibit approximately constant curvature and oscillatory torsion, from **softer folds**, for which both curvature and torsion oscillate [1206.0461].

The 2024 paper extends this mechanics viewpoint to **1DoF mechanisms**, **distributed actuation by spontaneous curvature**, and **cross-talk between multiple folds** [2412.18265]. For a curved fold line \(\bar{\gamma}(S)\) mapped isometrically to a space curve \(\gamma(s)\), the paper uses crease and surface frames to derive compatibility conditions and the relation
\[
\theta=\pi-2\alpha,
\]
so that the local fold opening is encoded by a scalar field \(\alpha(s)\) [2412.18265]. It states that the folded curvature satisfies a restriction of the form
\[
\kappa \ge \bar{\kappa},
\]
and identifies a 1-DoF mechanism when the rulings on the two sides are collinear in the flat state, implying \(\tau=0\) and an ODE for \(\alpha(s)\) [2412.18265]. The same paper compares a geometric approach with two mechanics-based actuation models—an active shell model with prescribed target curvature tensor and a 3D active elasticity model with prescribed target strain—and reports that both predict tapering near the crease due to incompatibility [2412.18265]. It also argues that synchronous folding of multiple curved folds is energetically favored over sequential folding [2412.18265].

Across these works, COrigami in the mechanics sense denotes a family of problems in which curvature is not merely decorative but arises from developability, crease geometry, fold stiffness, elastic frustration, and multi-fold coupling [1206.0461; 2102.03753; 2412.18265].

## 7. Conceptual synthesis and significance

The AI and mechanics meanings of COrigami occupy different technical strata, but they share a common research logic: origami design is constrained not only by intended shape but by rigorous geometric compatibility. In the AI pipeline, this appears as the refusal to let a language model freely emit long crease-pattern descriptions, replacing unconstrained generation with deterministic solvers for foldability-critical stages [2606.19177]. In curved-crease mechanics, it appears as the requirement that curvature, torsion, ruling directions, and developability satisfy nontrivial compatibility conditions [1206.0461; 2412.18265], or that emergent curvature must be understood through mechanical frustration and phase-structured deformation regimes [2102.03753].

The 2026 COrigami paper situates its contribution in a broader claim about structured creativity: AI works best when **semantic understanding** is delegated to language models, **hard constraints** to symbolic or geometric solvers, and **subjective quality** to a multimodal evaluator coupled to reinforcement learning [2606.19177]. The curved-crease literature, by contrast, shows that even seemingly simple fold networks can exhibit intricate couplings between local kinematics and global morphology [1206.0461; 2102.03753; 2412.18265]. This suggests a possible convergence between the two literatures, although such a synthesis is not claimed in the cited papers: AI-assisted origami systems may eventually need to incorporate richer mechanics if they are to move beyond zero-thickness flat-foldability toward physically faithful curved-shell design.

At present, the most precise usage of **COrigami** depends on context. In contemporary computational design, it names a specific AI-assisted pipeline for producing mathematically valid, visually meaningful origami blueprints from text [2606.19177]. In mechanics, it denotes curved origami structures whose geometry and actuation are governed by developability, stiffness, and elastic incompatibility [1206.0461; 2102.03753; 2412.18265]. The coexistence of these meanings reflects the breadth of origami research itself: a field spanning symbolic computation, geometric algorithms, differential geometry, elasticity, and co-creative design.

Source: https://www.emergentmind.com/topics/corigami