# Geometric loop generation on Torus T^2 L=2 (test)

Source: [Geometric loop generation on Torus T^2 L=2 (test)](<https://www.sota2.com/research/sota/geometric-loop-generation-on-torus-t-2-l-2-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-18T08:11:47.456Z

Evaluation of geometric loop generation performance on the test split of the Torus T^2 dataset variant with path length L=2, measured using Per-segment Chamfer distance.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Geometric loop generation |
| Dataset (including any stated split) | Torus T^2 L=2 (test) |
| Metric | Per-segment Chamfer Distance |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 5 of 5 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Per-segment Chamfer Distance](<https://www.sota2.com/research/sota/geometric-loop-generation-on-torus-t-2-l-2-test.md?metric=Per-segment%20Chamfer%20Distance>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Per-segment Chamfer Distance | Result date | Report paper |
| --- | --- | --- | --- |
| Transport | 1.68 | Not recorded | 69ba31ab0636e5bf4afc09ba |
| Homotopy | 1.68 | Not recorded | 69ba31ab0636e5bf4afc09ba |
| Transp. Attn. | 1.83 | Not recorded | 69ba31ab0636e5bf4afc09ba |
| Transf. (WC) | 1.89 | Not recorded | 69ba31ab0636e5bf4afc09ba |
| Cover | 2.15 | Not recorded | 69ba31ab0636e5bf4afc09ba |

### Transport

Per-segment Chamfer Distance: 1.68. Result ID: 69ba5e1f47f34ce924a41ca7.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31ab0636e5bf4afc09ba.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ba31ab0636e5bf4afc09ba.md>)

Recorded evaluation configuration: {"Type":"B"}. This record does not assert independent reproduction.

### Homotopy

Per-segment Chamfer Distance: 1.68. Result ID: 69ba5e1f47f34ce924a41ca8.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31ab0636e5bf4afc09ba.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ba31ab0636e5bf4afc09ba.md>)

Recorded evaluation configuration: {"Type":"B"}. This record does not assert independent reproduction.

### Transp. Attn.

Per-segment Chamfer Distance: 1.83. Result ID: 69ba5e1f47f34ce924a41ca6.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31ab0636e5bf4afc09ba.md>)

Recorded evaluation configuration: {"Type":"A"}. This record does not assert independent reproduction.

### Transf. (WC)

Per-segment Chamfer Distance: 1.89. Result ID: 69ba5e1f47f34ce924a41ca4.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31ab0636e5bf4afc09ba.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa703d1e209131cea49db1.md>)

[Implementation](<https://github.com/tensorflow/tensor2tensor>)

Recorded evaluation configuration: {"Type":"A"}. This record does not assert independent reproduction.

### Cover

Per-segment Chamfer Distance: 2.15. Result ID: 69ba5e1f47f34ce924a41ca5.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31ab0636e5bf4afc09ba.md>)

Recorded evaluation configuration: {"Type":"A"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.