# Long-horizon robotic manipulation on Real-world Unseen Lighting

Source: [Long-horizon robotic manipulation on Real-world Unseen Lighting](<https://www.sota2.com/research/sota/long-horizon-robotic-manipulation-on-real-world-unseen-lighting>)

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

Data updated: 2026-02-26T06:05:41.818Z

Evaluation of long-horizon robotic manipulation performance on the Real-world Unseen Lighting dataset, focusing on success rates across sequential steps (1 through 6) and the average trajectory length.

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 | Long-horizon robotic manipulation |
| Dataset (including any stated split) | Real-world Unseen Lighting |
| Metric | Success Rate (Step 1) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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

- [Success Rate (Step 1)](<https://www.sota2.com/research/sota/long-horizon-robotic-manipulation-on-real-world-unseen-lighting.md?metric=Success%20Rate%20(Step%201)>): Higher is better; unit: PERCENT.

- [Success Rate (Step 2)](<https://www.sota2.com/research/sota/long-horizon-robotic-manipulation-on-real-world-unseen-lighting.md?metric=Success%20Rate%20(Step%202)>): Higher is better; unit: PERCENT.

- [Success Rate (Step 3)](<https://www.sota2.com/research/sota/long-horizon-robotic-manipulation-on-real-world-unseen-lighting.md?metric=Success%20Rate%20(Step%203)>): Higher is better; unit: PERCENT.

- [Success Rate (Step 4)](<https://www.sota2.com/research/sota/long-horizon-robotic-manipulation-on-real-world-unseen-lighting.md?metric=Success%20Rate%20(Step%204)>): Higher is better; unit: PERCENT.

- [Success Rate (Step 5)](<https://www.sota2.com/research/sota/long-horizon-robotic-manipulation-on-real-world-unseen-lighting.md?metric=Success%20Rate%20(Step%205)>): Higher is better; unit: PERCENT.

- [Success Rate (Step 6)](<https://www.sota2.com/research/sota/long-horizon-robotic-manipulation-on-real-world-unseen-lighting.md?metric=Success%20Rate%20(Step%206)>): Higher is better; unit: PERCENT.

- [Average Length](<https://www.sota2.com/research/sota/long-horizon-robotic-manipulation-on-real-world-unseen-lighting.md?metric=Average%20Length>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Success Rate (Step 1) | Result date | Report paper |
| --- | --- | --- | --- |
| PALM | 80 | Not recorded | 6965d1ae6343fa78c832389c |
| OpenVLA | 55 | Not recorded | 6965d1ae6343fa78c832389c |
| Octo | 50 | Not recorded | 6965d1ae6343fa78c832389c |

### PALM

Success Rate (Step 1): 80. Result ID: 6999daa55698665ec459cf74.

[Reporting paper](<https://www.sota2.com/research/paper/6965d1ae6343fa78c832389c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6965d1ae6343fa78c832389c.md>)

Recorded evaluation configuration: {"fine-tuned":true}. This record does not assert independent reproduction.

### OpenVLA

Success Rate (Step 1): 55. Result ID: 6999daa55698665ec459cf72.

[Reporting paper](<https://www.sota2.com/research/paper/6965d1ae6343fa78c832389c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74c31e209131cea63a36.md>)

[Implementation](<https://openvla.github.io/>)

Recorded evaluation configuration: {"fine-tuned":true}. This record does not assert independent reproduction.

### Octo

Success Rate (Step 1): 50. Result ID: 6999daa55698665ec459cf73.

[Reporting paper](<https://www.sota2.com/research/paper/6965d1ae6343fa78c832389c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73551e209131cea5be6c.md>)

[Implementation](<https://octo-models.github.io/>)

Recorded evaluation configuration: {"fine-tuned":true}. This record does not assert independent reproduction.

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