# Phase Reconstruction on Synthetic LTEM Noisy per-sample evaluation (test)

Source: [Phase Reconstruction on Synthetic LTEM Noisy per-sample evaluation (test)](<https://www.sota2.com/research/sota/phase-reconstruction-on-synthetic-ltem-noisy-per-sample-evaluation-test>)

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

Data updated: 2026-04-30T07:52:46.476Z

Evaluation of Phase Reconstruction models using the Synthetic LTEM dataset under noisy conditions, specifically on the test split using per-sample metrics.

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 | Phase Reconstruction |
| Dataset (including any stated split) | Synthetic LTEM Noisy per-sample evaluation (test) |
| Metric | MSE (x10^-2) |
| 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

- [MSE (x10^-2)](<https://www.sota2.com/research/sota/phase-reconstruction-on-synthetic-ltem-noisy-per-sample-evaluation-test.md?metric=MSE%20(x10%5E-2)>): Lower is better; unit: SCALAR.

- [MAE (x10^-1)](<https://www.sota2.com/research/sota/phase-reconstruction-on-synthetic-ltem-noisy-per-sample-evaluation-test.md?metric=MAE%20(x10%5E-1)>): Lower is better; unit: SCALAR.

- [PSNR (dB)](<https://www.sota2.com/research/sota/phase-reconstruction-on-synthetic-ltem-noisy-per-sample-evaluation-test.md?metric=PSNR%20(dB)>): Higher is better; unit: SCALAR.

- [SSIM (%)](<https://www.sota2.com/research/sota/phase-reconstruction-on-synthetic-ltem-noisy-per-sample-evaluation-test.md?metric=SSIM%20(%25)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | MSE (x10^-2) | Result date | Report paper |
| --- | --- | --- | --- |
| Ours | 0.95 | Not recorded | 69f2e20655a3c32af89fc5d8 |
| PPN | 1.21 | Not recorded | 69f2e20655a3c32af89fc5d8 |
| DPI | 1.47 | Not recorded | 69f2e20655a3c32af89fc5d8 |
| PtychoNN | 2.1 | Not recorded | 69f2e20655a3c32af89fc5d8 |
| PtyNet | 12.53 | Not recorded | 69f2e20655a3c32af89fc5d8 |

### Ours

MSE (x10^-2): 0.95. Result ID: 69f30a0647f34ce924d937ad.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e20655a3c32af89fc5d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f2e20655a3c32af89fc5d8.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### PPN

MSE (x10^-2): 1.21. Result ID: 69f30a0647f34ce924d937ac.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e20655a3c32af89fc5d8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa4a1f1e209131ce93bfb7.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### DPI

MSE (x10^-2): 1.47. Result ID: 69f30a0647f34ce924d937aa.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e20655a3c32af89fc5d8.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### PtychoNN

MSE (x10^-2): 2.1. Result ID: 69f30a0647f34ce924d937a9.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e20655a3c32af89fc5d8.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### PtyNet

MSE (x10^-2): 12.53. Result ID: 69f30a0647f34ce924d937ab.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e20655a3c32af89fc5d8.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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