# Free-form answer generation on KLAR (OOD)

Source: [Free-form answer generation on KLAR (OOD)](<https://www.sota2.com/research/sota/free-form-answer-generation-on-klar-ood>)

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

Data updated: 2026-06-08T06:32:02.796Z

Evaluation of model performance on the KLAR (OOD) dataset for the task of free-form answer generation, measured using accuracy.

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 | Free-form answer generation |
| Dataset (including any stated split) | KLAR (OOD) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 12 of 12 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/free-form-answer-generation-on-klar-ood.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| GRPO | 38.13 | Not recorded | 6a264c8590966ade919be956 |
| Baseline | 35.76 | Not recorded | 6a264c8590966ade919be956 |
| CPT + GRPO | 30.75 | Not recorded | 6a264c8590966ade919be956 |
| CPT + SFT | 30.21 | Not recorded | 6a264c8590966ade919be956 |
| SFT | 29.96 | Not recorded | 6a264c8590966ade919be956 |
| CPT | 23.75 | Not recorded | 6a264c8590966ade919be956 |
| CPT + GRPO | 17.6 | Not recorded | 6a264c8590966ade919be956 |
| GRPO | 16.7 | Not recorded | 6a264c8590966ade919be956 |
| Baseline | 13.3 | Not recorded | 6a264c8590966ade919be956 |
| CPT + SFT | 8.6 | Not recorded | 6a264c8590966ade919be956 |
| CPT | 8.3 | Not recorded | 6a264c8590966ade919be956 |
| SFT | 7.8 | Not recorded | 6a264c8590966ade919be956 |

### GRPO

Accuracy: 38.13. Result ID: 6a266179fbc373592c226c5a.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-2.5-7B"}. This record does not assert independent reproduction.

### Baseline

Accuracy: 35.76. Result ID: 6a266179fbc373592c226c56.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-2.5-7B"}. This record does not assert independent reproduction.

### CPT + GRPO

Accuracy: 30.75. Result ID: 6a266179fbc373592c226c5b.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-2.5-7B"}. This record does not assert independent reproduction.

### CPT + SFT

Accuracy: 30.21. Result ID: 6a266179fbc373592c226c59.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-2.5-7B"}. This record does not assert independent reproduction.

### SFT

Accuracy: 29.96. Result ID: 6a266179fbc373592c226c58.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-2.5-7B"}. This record does not assert independent reproduction.

### CPT

Accuracy: 23.75. Result ID: 6a266179fbc373592c226c57.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-2.5-7B"}. This record does not assert independent reproduction.

### CPT + GRPO

Accuracy: 17.6. Result ID: 6a266179fbc373592c226c55.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"OLMo-2-1124-7B"}. This record does not assert independent reproduction.

### GRPO

Accuracy: 16.7. Result ID: 6a266179fbc373592c226c54.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"OLMo-2-1124-7B"}. This record does not assert independent reproduction.

### Baseline

Accuracy: 13.3. Result ID: 6a266179fbc373592c226c50.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"OLMo-2-1124-7B"}. This record does not assert independent reproduction.

### CPT + SFT

Accuracy: 8.6. Result ID: 6a266179fbc373592c226c53.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"OLMo-2-1124-7B"}. This record does not assert independent reproduction.

### CPT

Accuracy: 8.3. Result ID: 6a266179fbc373592c226c51.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"OLMo-2-1124-7B"}. This record does not assert independent reproduction.

### SFT

Accuracy: 7.8. Result ID: 6a266179fbc373592c226c52.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"OLMo-2-1124-7B"}. This record does not assert independent reproduction.

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