# Knowledge Base Question Answering on DISCOS KB (val)

Source: [Knowledge Base Question Answering on DISCOS KB (val)](<https://www.sota2.com/research/sota/knowledge-base-question-answering-on-discos-kb-val>)

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

Data updated: 2026-02-26T05:34:16.454Z

Evaluation of Knowledge Base Question Answering performance on the validation split of the DISCOS KB dataset, reporting accuracies across functions, entities, relations, concepts, and attributes.

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 | Knowledge Base Question Answering |
| Dataset (including any stated split) | DISCOS KB (val) |
| Metric | Accuracy (Functions) |
| 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

- [Accuracy (Functions)](<https://www.sota2.com/research/sota/knowledge-base-question-answering-on-discos-kb-val.md?metric=Accuracy%20(Functions)>): Higher is better; unit: PERCENT.

- [Accuracy (Entities)](<https://www.sota2.com/research/sota/knowledge-base-question-answering-on-discos-kb-val.md?metric=Accuracy%20(Entities)>): Higher is better; unit: PERCENT.

- [Accuracy (Relations)](<https://www.sota2.com/research/sota/knowledge-base-question-answering-on-discos-kb-val.md?metric=Accuracy%20(Relations)>): Higher is better; unit: PERCENT.

- [Accuracy (Concepts)](<https://www.sota2.com/research/sota/knowledge-base-question-answering-on-discos-kb-val.md?metric=Accuracy%20(Concepts)>): Higher is better; unit: PERCENT.

- [Accuracy (Attributes)](<https://www.sota2.com/research/sota/knowledge-base-question-answering-on-discos-kb-val.md?metric=Accuracy%20(Attributes)>): Higher is better; unit: PERCENT.

- [Overall Accuracy](<https://www.sota2.com/research/sota/knowledge-base-question-answering-on-discos-kb-val.md?metric=Overall%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (Functions) | Result date | Report paper |
| --- | --- | --- | --- |
| CosmicRoBERTa | 79 | Not recorded | 68fa6f991e209131cea46874 |
| GPT-4 | 79 | Not recorded | 68fa6f991e209131cea46874 |
| ChatGPT-3.5 | 51.6 | Not recorded | 68fa6f991e209131cea46874 |

### CosmicRoBERTa

Accuracy (Functions): 79. Result ID: 6990ceda3a6c314849037146.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f991e209131cea46874.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f991e209131cea46874.md>)

[Implementation](<https://github.com/PaulDrm/DISCOSQA>)

Recorded evaluation configuration: {"Model architecture":"Domain-specific retrained encoder"}. This record does not assert independent reproduction.

### GPT-4

Accuracy (Functions): 79. Result ID: 6990ceda3a6c314849037147.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f991e209131cea46874.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa727a1e209131cea56a1e.md>)

[Implementation](<https://github.com/openai/evals>)

Recorded evaluation configuration: {"Prompting":"Few-shot until context limit reached"}. This record does not assert independent reproduction.

### ChatGPT-3.5

Accuracy (Functions): 51.6. Result ID: 6990ceda3a6c314849037148.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f991e209131cea46874.md>)

Recorded evaluation configuration: {"Prompting":"Few-shot until context limit reached"}. This record does not assert independent reproduction.

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