# Logical Refinement of Natural Language Explanations on WorldTree

Source: [Logical Refinement of Natural Language Explanations on WorldTree](<https://www.sota2.com/research/sota/logical-refinement-of-natural-language-explanations-on-worldtree>)

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

Data updated: 2026-02-26T05:47:13.104Z

Evaluation of model performance on iteratively refining natural language explanations using the WorldTree dataset, tracking initial performance, final performance, and the number of iterations/calls required.

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 | Logical Refinement of Natural Language Explanations |
| Dataset (including any stated split) | WorldTree |
| Metric | Initial Performance |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [Initial Performance](<https://www.sota2.com/research/sota/logical-refinement-of-natural-language-explanations-on-worldtree.md?metric=Initial%20Performance>): Higher is better; unit: PERCENT.

- [Final Performance](<https://www.sota2.com/research/sota/logical-refinement-of-natural-language-explanations-on-worldtree.md?metric=Final%20Performance>): Higher is better; unit: PERCENT.

- [Iterations Count](<https://www.sota2.com/research/sota/logical-refinement-of-natural-language-explanations-on-worldtree.md?metric=Iterations%20Count>): Lower is better; unit: SCALAR.

- [API Calls Count](<https://www.sota2.com/research/sota/logical-refinement-of-natural-language-explanations-on-worldtree.md?metric=API%20Calls%20Count>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Initial Performance | Result date | Report paper |
| --- | --- | --- | --- |
| Faithful-Refiner | 9 | Not recorded | 68fa78e71e209131cea7b1e8 |
| Faithful-Refiner | 7 | Not recorded | 68fa78e71e209131cea7b1e8 |
| Faithful-Refiner | 6 | Not recorded | 68fa78e71e209131cea7b1e8 |
| Faithful-Refiner | 5 | Not recorded | 68fa78e71e209131cea7b1e8 |
| Explanation-Refiner | 3 | Not recorded | 68fa78e71e209131cea7b1e8 |
| Explanation-Refiner | 2 | Not recorded | 68fa78e71e209131cea7b1e8 |
| Explanation-Refiner | 2 | Not recorded | 68fa78e71e209131cea7b1e8 |
| Explanation-Refiner | 0 | Not recorded | 68fa78e71e209131cea7b1e8 |

### Faithful-Refiner

Initial Performance: 9. Result ID: 6995ac7a3badfaffe6ab286c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e71e209131cea7b1e8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78e71e209131cea7b1e8.md>)

[Implementation](<https://github.com/neurosymbolicai/faithful_and_robust_nli_refinement>)

Recorded evaluation configuration: {"LLM":"GPT-4o"}. This record does not assert independent reproduction.

### Faithful-Refiner

Initial Performance: 7. Result ID: 6995ac7a3badfaffe6ab286d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e71e209131cea7b1e8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78e71e209131cea7b1e8.md>)

[Implementation](<https://github.com/neurosymbolicai/faithful_and_robust_nli_refinement>)

Recorded evaluation configuration: {"LLM":"Deepseek-V3"}. This record does not assert independent reproduction.

### Faithful-Refiner

Initial Performance: 6. Result ID: 6995ac7a3badfaffe6ab286a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e71e209131cea7b1e8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78e71e209131cea7b1e8.md>)

[Implementation](<https://github.com/neurosymbolicai/faithful_and_robust_nli_refinement>)

Recorded evaluation configuration: {"LLM":"Llama3.1-70b"}. This record does not assert independent reproduction.

### Faithful-Refiner

Initial Performance: 5. Result ID: 6995ac7a3badfaffe6ab286b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e71e209131cea7b1e8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78e71e209131cea7b1e8.md>)

[Implementation](<https://github.com/neurosymbolicai/faithful_and_robust_nli_refinement>)

Recorded evaluation configuration: {"LLM":"GPT-4o-mini"}. This record does not assert independent reproduction.

### Explanation-Refiner

Initial Performance: 3. Result ID: 6995ac7a3badfaffe6ab2869.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e71e209131cea7b1e8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa752f1e209131cea66324.md>)

Recorded evaluation configuration: {"LLM":"Deepseek-V3"}. This record does not assert independent reproduction.

### Explanation-Refiner

Initial Performance: 2. Result ID: 6995ac7a3badfaffe6ab2866.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e71e209131cea7b1e8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa752f1e209131cea66324.md>)

Recorded evaluation configuration: {"LLM":"Llama3.1-70b"}. This record does not assert independent reproduction.

### Explanation-Refiner

Initial Performance: 2. Result ID: 6995ac7a3badfaffe6ab2868.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e71e209131cea7b1e8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa752f1e209131cea66324.md>)

Recorded evaluation configuration: {"LLM":"GPT-4o"}. This record does not assert independent reproduction.

### Explanation-Refiner

Initial Performance: 0. Result ID: 6995ac7a3badfaffe6ab2867.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e71e209131cea7b1e8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa752f1e209131cea66324.md>)

Recorded evaluation configuration: {"LLM":"GPT-4o-mini"}. This record does not assert independent reproduction.

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