# Multi-hop QA on CompWebQ

Source: [Multi-hop QA on CompWebQ](<https://www.sota2.com/research/sota/multi-hop-qa-on-compwebq>)

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

Data updated: 2026-02-26T05:31:10.316Z

Evaluation of model performance on the CompWebQ dataset for multi-hop question answering, typically 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 | Multi-hop QA |
| Dataset (including any stated split) | CompWebQ |
| Metric | Accuracy |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/multi-hop-qa-on-compwebq.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| CBR | 70.4 | Not recorded | 69438a74423ea422657293a2 |
| GPT-4+RFKG-CoT | 65.1 | Not recorded | 69438a74423ea422657293a2 |
| GPT-4+KG-CoT | 62.3 | Not recorded | 69438a74423ea422657293a2 |
| ChatGPT+RFKG-CoT | 61.4 | Not recorded | 69438a74423ea422657293a2 |
| UniKGQA | 52.2 | Not recorded | 69438a74423ea422657293a2 |
| ChatGPT+KG-CoT | 51.6 | Not recorded | 69438a74423ea422657293a2 |
| ChatGPT+SC | 45.4 | Not recorded | 69438a74423ea422657293a2 |
| ChatGPT+CoT | 38.8 | Not recorded | 69438a74423ea422657293a2 |
| ChatGPT+IO | 37.6 | Not recorded | 69438a74423ea422657293a2 |

### CBR

Accuracy: 70.4. Result ID: 6996bced5698665ec407cf7d.

[Reporting paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b331e209131cea2ccf9.md>)

Recorded evaluation configuration: {"AccessKB":true}. This record does not assert independent reproduction.

### GPT-4+RFKG-CoT

Accuracy: 65.1. Result ID: 6996bced5698665ec407cf82.

[Reporting paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

Recorded evaluation configuration: {"AccessKB":true,"Backbone":"GPT-4"}. This record does not assert independent reproduction.

### GPT-4+KG-CoT

Accuracy: 62.3. Result ID: 6996bced5698665ec407cf80.

[Reporting paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

Recorded evaluation configuration: {"AccessKB":true,"Backbone":"GPT-4"}. This record does not assert independent reproduction.

### ChatGPT+RFKG-CoT

Accuracy: 61.4. Result ID: 6996bced5698665ec407cf81.

[Reporting paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

Recorded evaluation configuration: {"AccessKB":true,"Backbone":"ChatGPT"}. This record does not assert independent reproduction.

### UniKGQA

Accuracy: 52.2. Result ID: 6996bced5698665ec407cf7e.

[Reporting paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ea71e209131cea411f3.md>)

Recorded evaluation configuration: {"AccessKB":true}. This record does not assert independent reproduction.

### ChatGPT+KG-CoT

Accuracy: 51.6. Result ID: 6996bced5698665ec407cf7f.

[Reporting paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

Recorded evaluation configuration: {"AccessKB":true,"Backbone":"ChatGPT"}. This record does not assert independent reproduction.

### ChatGPT+SC

Accuracy: 45.4. Result ID: 6996bced5698665ec407cf7c.

[Reporting paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb61e209131cea41b33.md>)

Recorded evaluation configuration: {"AccessKB":false,"Backbone":"ChatGPT"}. This record does not assert independent reproduction.

### ChatGPT+CoT

Accuracy: 38.8. Result ID: 6996bced5698665ec407cf7b.

[Reporting paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e571e209131cea3f27b.md>)

Recorded evaluation configuration: {"AccessKB":false,"Backbone":"ChatGPT"}. This record does not assert independent reproduction.

### ChatGPT+IO

Accuracy: 37.6. Result ID: 6996bced5698665ec407cf7a.

[Reporting paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e7e1e209131cea404d4.md>)

Recorded evaluation configuration: {"AccessKB":false,"Backbone":"ChatGPT"}. This record does not assert independent reproduction.

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