# Knowledge Graph Question Answering on WebQSP (Performance and Cost Analysis)

Source: [Knowledge Graph Question Answering on WebQSP (Performance and Cost Analysis)](<https://www.sota2.com/research/sota/knowledge-graph-question-answering-on-webqsp-performance-and-cost-analysis>)

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

Data updated: 2026-03-23T11:45:37.630Z

Evaluation of model performance on the WebQSP dataset for Knowledge Graph Question Answering, reporting standard accuracy metrics (Hits@1, F1) alongside operational metrics related to LLM usage and cost.

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 Graph Question Answering |
| Dataset (including any stated split) | WebQSP |
| Metric | Hits@1 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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

- [Hits@1](<https://www.sota2.com/research/sota/knowledge-graph-question-answering-on-webqsp-performance-and-cost-analysis.md?metric=Hits%401>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/knowledge-graph-question-answering-on-webqsp-performance-and-cost-analysis.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

- [Avg. LLM Calls](<https://www.sota2.com/research/sota/knowledge-graph-question-answering-on-webqsp-performance-and-cost-analysis.md?metric=Avg.%20LLM%20Calls>): Lower is better; unit: SCALAR.

- [Avg. Tokens](<https://www.sota2.com/research/sota/knowledge-graph-question-answering-on-webqsp-performance-and-cost-analysis.md?metric=Avg.%20Tokens>): Lower is better; unit: SCALAR.

- [Avg. Cost (USD) / 10K Qs](<https://www.sota2.com/research/sota/knowledge-graph-question-answering-on-webqsp-performance-and-cost-analysis.md?metric=Avg.%20Cost%20(USD)%20%2F%2010K%20Qs>): Lower is better; unit: OTHER.

## Recorded results

| Method | Hits@1 | Result date | Report paper |
| --- | --- | --- | --- |
| GCR | 91.68 | Not recorded | 69c0c8ef0636e5bf4afc3ac8 |
| RouterKGQA (Full) | 91.15 | Not recorded | 69c0c8ef0636e5bf4afc3ac8 |
| GNN-RAG | 90.73 | Not recorded | 69c0c8ef0636e5bf4afc3ac8 |
| MemQ | 88.74 | Not recorded | 69c0c8ef0636e5bf4afc3ac8 |
| ChatKBQA | 86.33 | Not recorded | 69c0c8ef0636e5bf4afc3ac8 |
| RoG | 86.18 | Not recorded | 69c0c8ef0636e5bf4afc3ac8 |
| RouterKGQA (Stage 2+3) | 86.07 | Not recorded | 69c0c8ef0636e5bf4afc3ac8 |
| PoG | 83.13 | Not recorded | 69c0c8ef0636e5bf4afc3ac8 |
| ToG | 77.2 | Not recorded | 69c0c8ef0636e5bf4afc3ac8 |

### GCR

Hits@1: 91.68. Result ID: 69c127d747f34ce924a9b8dc.

[Reporting paper](<https://www.sota2.com/research/paper/69c0c8ef0636e5bf4afc3ac8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78d21e209131cea7a9a7.md>)

Recorded evaluation configuration: {"Category":"Retrieval-based"}. This record does not assert independent reproduction.

### RouterKGQA (Full)

Hits@1: 91.15. Result ID: 69c127d747f34ce924a9b8e1.

[Reporting paper](<https://www.sota2.com/research/paper/69c0c8ef0636e5bf4afc3ac8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c0c8ef0636e5bf4afc3ac8.md>)

Recorded evaluation configuration: {"Category":"Ours"}. This record does not assert independent reproduction.

### GNN-RAG

Hits@1: 90.73. Result ID: 69c127d747f34ce924a9b8d9.

[Reporting paper](<https://www.sota2.com/research/paper/69c0c8ef0636e5bf4afc3ac8.md>)

Recorded evaluation configuration: {"Category":"Retrieval-based"}. This record does not assert independent reproduction.

### MemQ

Hits@1: 88.74. Result ID: 69c127d747f34ce924a9b8dd.

[Reporting paper](<https://www.sota2.com/research/paper/69c0c8ef0636e5bf4afc3ac8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77471e209131cea72344.md>)

Recorded evaluation configuration: {"Category":"Retrieval-based"}. This record does not assert independent reproduction.

### ChatKBQA

Hits@1: 86.33. Result ID: 69c127d747f34ce924a9b8db.

[Reporting paper](<https://www.sota2.com/research/paper/69c0c8ef0636e5bf4afc3ac8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa757a1e209131cea683b7.md>)

[Implementation](<https://github.com/LHRLAB/ChatKBQA>)

Recorded evaluation configuration: {"Category":"Retrieval-based"}. This record does not assert independent reproduction.

### RoG

Hits@1: 86.18. Result ID: 69c127d747f34ce924a9b8da.

[Reporting paper](<https://www.sota2.com/research/paper/69c0c8ef0636e5bf4afc3ac8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72591e209131cea55d71.md>)

Recorded evaluation configuration: {"Category":"Retrieval-based"}. This record does not assert independent reproduction.

### RouterKGQA (Stage 2+3)

Hits@1: 86.07. Result ID: 69c127d747f34ce924a9b8e0.

[Reporting paper](<https://www.sota2.com/research/paper/69c0c8ef0636e5bf4afc3ac8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c0c8ef0636e5bf4afc3ac8.md>)

Recorded evaluation configuration: {"Category":"Ours"}. This record does not assert independent reproduction.

### PoG

Hits@1: 83.13. Result ID: 69c127d747f34ce924a9b8df.

[Reporting paper](<https://www.sota2.com/research/paper/69c0c8ef0636e5bf4afc3ac8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75741e209131cea68054.md>)

[Implementation](<https://github.com/liyichen-cly/PoG>)

Recorded evaluation configuration: {"Category":"Agent-based"}. This record does not assert independent reproduction.

### ToG

Hits@1: 77.2. Result ID: 69c127d747f34ce924a9b8de.

[Reporting paper](<https://www.sota2.com/research/paper/69c0c8ef0636e5bf4afc3ac8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72a21e209131cea57b26.md>)

Recorded evaluation configuration: {"Category":"Agent-based"}. This record does not assert independent reproduction.

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