# Explanation Generation on VariErr

Source: [Explanation Generation on VariErr](<https://www.sota2.com/research/sota/explanation-generation-on-varierr>)

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

Data updated: 2026-05-28T10:35:53.004Z

Evaluation of model performance on generating explanations using the VariErr dataset, assessed using feature divergence metrics such as SFT (Mean Feature KL) and CAPO (Mean Feature KL).

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 | Explanation Generation |
| Dataset (including any stated split) | VariErr |
| Metric | SFT (Mean Feature KL) |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 2 of 2 |
| 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

- [SFT (Mean Feature KL)](<https://www.sota2.com/research/sota/explanation-generation-on-varierr.md?metric=SFT%20(Mean%20Feature%20KL)>): Lower is better; unit: SCALAR.

- [CAPO (Mean Feature KL)](<https://www.sota2.com/research/sota/explanation-generation-on-varierr.md?metric=CAPO%20(Mean%20Feature%20KL)>): Lower is better; unit: SCALAR.

- [Relative Change](<https://www.sota2.com/research/sota/explanation-generation-on-varierr.md?metric=Relative%20Change>): Higher is better; unit: PERCENT.

## Recorded results

| Method | SFT (Mean Feature KL) | Result date | Report paper |
| --- | --- | --- | --- |
| Qwen | 0.0839 | Not recorded | 6a17cc37dbd7150506f8146e |
| Llama | 0.1131 | Not recorded | 6a17cc37dbd7150506f8146e |

### Qwen

SFT (Mean Feature KL): 0.0839. Result ID: 6a181a83fbc373592c129a22.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc37dbd7150506f8146e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Model":"Qwen"}. This record does not assert independent reproduction.

### Llama

SFT (Mean Feature KL): 0.1131. Result ID: 6a181a83fbc373592c129a21.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc37dbd7150506f8146e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Model":"Llama"}. This record does not assert independent reproduction.

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