# Multi-hop Question Answering on MQuAKE

Source: [Multi-hop Question Answering on MQuAKE](<https://www.sota2.com/research/sota/multi-hop-question-answering-on-mquake>)

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

Data updated: 2026-02-26T05:49:51.014Z

Evaluation of model performance on the MQuAKE dataset for multi-hop question answering, typically reporting MHQ 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 Question Answering |
| Dataset (including any stated split) | MQuAKE |
| Metric | MHQ Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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

- [MHQ Accuracy](<https://www.sota2.com/research/sota/multi-hop-question-answering-on-mquake.md?metric=MHQ%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | MHQ Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Redundant Editing (ROME) | 31.6 | Not recorded | 69608b6453fdd5c361744131 |
| Redundant Editing (ROME) | 23.9 | Not recorded | 69608b6453fdd5c361744131 |
| Redundant Editing (ROME) | 20.4 | Not recorded | 69608b6453fdd5c361744131 |
| Redundant Editing (ROME) | 20.2 | Not recorded | 69608b6453fdd5c361744131 |
| Redundant Editing (ROME) | 17.3 | Not recorded | 69608b6453fdd5c361744131 |
| Redundant Editing (ROME) | 17 | Not recorded | 69608b6453fdd5c361744131 |
| ROME | 16.1 | Not recorded | 69608b6453fdd5c361744131 |
| ROME | 14.6 | Not recorded | 69608b6453fdd5c361744131 |
| ROME | 11.6 | Not recorded | 69608b6453fdd5c361744131 |
| ROME | 6.7 | Not recorded | 69608b6453fdd5c361744131 |

### Redundant Editing (ROME)

MHQ Accuracy: 31.6. Result ID: 699698ed5698665ec4e2857e.

[Reporting paper](<https://www.sota2.com/research/paper/69608b6453fdd5c361744131.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69608b6453fdd5c361744131.md>)

Recorded evaluation configuration: {"Layer(s)":"5,8,11,15,17,20"}. This record does not assert independent reproduction.

### Redundant Editing (ROME)

MHQ Accuracy: 23.9. Result ID: 699698ed5698665ec4e2857d.

[Reporting paper](<https://www.sota2.com/research/paper/69608b6453fdd5c361744131.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69608b6453fdd5c361744131.md>)

Recorded evaluation configuration: {"Layer(s)":"5,9,13,17,20"}. This record does not assert independent reproduction.

### Redundant Editing (ROME)

MHQ Accuracy: 20.4. Result ID: 699698ed5698665ec4e2857c.

[Reporting paper](<https://www.sota2.com/research/paper/69608b6453fdd5c361744131.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69608b6453fdd5c361744131.md>)

Recorded evaluation configuration: {"Layer(s)":"5,10,15,20"}. This record does not assert independent reproduction.

### Redundant Editing (ROME)

MHQ Accuracy: 20.2. Result ID: 699698ed5698665ec4e2857b.

[Reporting paper](<https://www.sota2.com/research/paper/69608b6453fdd5c361744131.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69608b6453fdd5c361744131.md>)

Recorded evaluation configuration: {"Layer(s)":"5,10,20"}. This record does not assert independent reproduction.

### Redundant Editing (ROME)

MHQ Accuracy: 17.3. Result ID: 699698ed5698665ec4e28579.

[Reporting paper](<https://www.sota2.com/research/paper/69608b6453fdd5c361744131.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69608b6453fdd5c361744131.md>)

Recorded evaluation configuration: {"Layer(s)":"5,15"}. This record does not assert independent reproduction.

### Redundant Editing (ROME)

MHQ Accuracy: 17. Result ID: 699698ed5698665ec4e2857a.

[Reporting paper](<https://www.sota2.com/research/paper/69608b6453fdd5c361744131.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69608b6453fdd5c361744131.md>)

Recorded evaluation configuration: {"Layer(s)":"5,20"}. This record does not assert independent reproduction.

### ROME

MHQ Accuracy: 16.1. Result ID: 699698ed5698665ec4e28575.

[Reporting paper](<https://www.sota2.com/research/paper/69608b6453fdd5c361744131.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e591e209131cea3f3c9.md>)

[Implementation](<https://rome.baulab.info/>)

Recorded evaluation configuration: {"Layer(s)":"5"}. This record does not assert independent reproduction.

### ROME

MHQ Accuracy: 14.6. Result ID: 699698ed5698665ec4e28576.

[Reporting paper](<https://www.sota2.com/research/paper/69608b6453fdd5c361744131.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e591e209131cea3f3c9.md>)

[Implementation](<https://rome.baulab.info/>)

Recorded evaluation configuration: {"Layer(s)":"10"}. This record does not assert independent reproduction.

### ROME

MHQ Accuracy: 11.6. Result ID: 699698ed5698665ec4e28577.

[Reporting paper](<https://www.sota2.com/research/paper/69608b6453fdd5c361744131.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e591e209131cea3f3c9.md>)

[Implementation](<https://rome.baulab.info/>)

Recorded evaluation configuration: {"Layer(s)":"15"}. This record does not assert independent reproduction.

### ROME

MHQ Accuracy: 6.7. Result ID: 699698ed5698665ec4e28578.

[Reporting paper](<https://www.sota2.com/research/paper/69608b6453fdd5c361744131.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e591e209131cea3f3c9.md>)

[Implementation](<https://rome.baulab.info/>)

Recorded evaluation configuration: {"Layer(s)":"20"}. This record does not assert independent reproduction.

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