# Long-Horizon Multi-Task Language Control on CALVIN ABC→D (test)

Source: [Long-Horizon Multi-Task Language Control on CALVIN ABC→D (test)](<https://www.sota2.com/research/sota/long-horizon-multi-task-language-control-on-calvin-abc-d-test>)

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

Data updated: 2026-02-26T06:04:17.501Z

Evaluation of model performance on the CALVIN ABC→D test set for long-horizon, multi-task language control. Performance is measured by sequential success rates (completing 1 through 5 tasks in a row) and the average length of successful sequences.

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 Multi-Task Language Control |
| Dataset (including any stated split) | CALVIN ABC→D (test) |
| Metric | Seq Success (1) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 13 of 13 |
| 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

- [Seq Success (1)](<https://www.sota2.com/research/sota/long-horizon-multi-task-language-control-on-calvin-abc-d-test.md?metric=Seq%20Success%20(1)>): Higher is better; unit: PERCENT.

- [Seq Success (2)](<https://www.sota2.com/research/sota/long-horizon-multi-task-language-control-on-calvin-abc-d-test.md?metric=Seq%20Success%20(2)>): Higher is better; unit: PERCENT.

- [Seq Success (3)](<https://www.sota2.com/research/sota/long-horizon-multi-task-language-control-on-calvin-abc-d-test.md?metric=Seq%20Success%20(3)>): Higher is better; unit: PERCENT.

- [Seq Success (4)](<https://www.sota2.com/research/sota/long-horizon-multi-task-language-control-on-calvin-abc-d-test.md?metric=Seq%20Success%20(4)>): Higher is better; unit: PERCENT.

- [Seq Success (5)](<https://www.sota2.com/research/sota/long-horizon-multi-task-language-control-on-calvin-abc-d-test.md?metric=Seq%20Success%20(5)>): Higher is better; unit: PERCENT.

- [Avg Sequence Length](<https://www.sota2.com/research/sota/long-horizon-multi-task-language-control-on-calvin-abc-d-test.md?metric=Avg%20Sequence%20Length>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Seq Success (1) | Result date | Report paper |
| --- | --- | --- | --- |
| CLOVER | 96 | Not recorded | 69857565c97a0b3e3b5d5ae6 |
| ReconVLA | 95.6 | Not recorded | 69857565c97a0b3e3b5d5ae6 |
| UniVLA | 95.5 | Not recorded | 69857565c97a0b3e3b5d5ae6 |
| VISTA-OFT | 94.8 | Not recorded | 69857565c97a0b3e3b5d5ae6 |
| DITA | 94.5 | Not recorded | 69857565c97a0b3e3b5d5ae6 |
| RoboDual | 94.4 | Not recorded | 69857565c97a0b3e3b5d5ae6 |
| VISTA | 93.5 | Not recorded | 69857565c97a0b3e3b5d5ae6 |
| RoboVLMs | 93.1 | Not recorded | 69857565c97a0b3e3b5d5ae6 |
| OpenVLA | 92.8 | Not recorded | 69857565c97a0b3e3b5d5ae6 |
| OpenVLA-OFT | 92.1 | Not recorded | 69857565c97a0b3e3b5d5ae6 |
| VPP | 90.9 | Not recorded | 69857565c97a0b3e3b5d5ae6 |
| SuSIE | 87 | Not recorded | 69857565c97a0b3e3b5d5ae6 |
| GR-1 | 85.4 | Not recorded | 69857565c97a0b3e3b5d5ae6 |

### CLOVER

Seq Success (1): 96. Result ID: 699742155698665ec492e450.

[Reporting paper](<https://www.sota2.com/research/paper/69857565c97a0b3e3b5d5ae6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa753e1e209131cea66ba3.md>)

[Implementation](<https://github.com/OpenDriveLab/CLOVER>)

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

### ReconVLA

Seq Success (1): 95.6. Result ID: 699742155698665ec492e454.

[Reporting paper](<https://www.sota2.com/research/paper/69857565c97a0b3e3b5d5ae6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7aaa1e209131cea84804.md>)

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

### UniVLA

Seq Success (1): 95.5. Result ID: 699742155698665ec492e453.

[Reporting paper](<https://www.sota2.com/research/paper/69857565c97a0b3e3b5d5ae6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78851e209131cea78d36.md>)

[Implementation](<https://github.com/OpenDriveLab/UniVLA>)

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

### VISTA-OFT

Seq Success (1): 94.8. Result ID: 699742155698665ec492e458.

[Reporting paper](<https://www.sota2.com/research/paper/69857565c97a0b3e3b5d5ae6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69857565c97a0b3e3b5d5ae6.md>)

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

### DITA

Seq Success (1): 94.5. Result ID: 699742155698665ec492e44f.

[Reporting paper](<https://www.sota2.com/research/paper/69857565c97a0b3e3b5d5ae6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b201e209131cea86b67.md>)

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

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

### RoboDual

Seq Success (1): 94.4. Result ID: 699742155698665ec492e451.

[Reporting paper](<https://www.sota2.com/research/paper/69857565c97a0b3e3b5d5ae6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76e91e209131cea6fe1c.md>)

[Implementation](<https://opendrivelab.com/RoboDual/>)

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

### VISTA

Seq Success (1): 93.5. Result ID: 699742155698665ec492e456.

[Reporting paper](<https://www.sota2.com/research/paper/69857565c97a0b3e3b5d5ae6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69857565c97a0b3e3b5d5ae6.md>)

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

### RoboVLMs

Seq Success (1): 93.1. Result ID: 699742155698665ec492e452.

[Reporting paper](<https://www.sota2.com/research/paper/69857565c97a0b3e3b5d5ae6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76601e209131cea6cbc6.md>)

[Implementation](<https://www.sota2.com/robovlms.github.io>)

Recorded evaluation configuration: {"View":"Single static third-person view"}. This record does not assert independent reproduction.

### OpenVLA

Seq Success (1): 92.8. Result ID: 699742155698665ec492e455.

[Reporting paper](<https://www.sota2.com/research/paper/69857565c97a0b3e3b5d5ae6.md>)

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

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

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

### OpenVLA-OFT

Seq Success (1): 92.1. Result ID: 699742155698665ec492e457.

[Reporting paper](<https://www.sota2.com/research/paper/69857565c97a0b3e3b5d5ae6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78471e209131cea7794b.md>)

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

### VPP

Seq Success (1): 90.9. Result ID: 699742155698665ec492e44e.

[Reporting paper](<https://www.sota2.com/research/paper/69857565c97a0b3e3b5d5ae6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78581e209131cea77da7.md>)

[Implementation](<https://video-prediction-policy.github.io/>)

Recorded evaluation configuration: {"View":"Single static third-person view"}. This record does not assert independent reproduction.

### SuSIE

Seq Success (1): 87. Result ID: 699742155698665ec492e44c.

[Reporting paper](<https://www.sota2.com/research/paper/69857565c97a0b3e3b5d5ae6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70ec1e209131cea4e08c.md>)

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

### GR-1

Seq Success (1): 85.4. Result ID: 699742155698665ec492e44d.

[Reporting paper](<https://www.sota2.com/research/paper/69857565c97a0b3e3b5d5ae6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71861e209131cea51c7f.md>)

[Implementation](<https://gr1-manipulation.github.io/>)

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.
