# End-to-end Question Answering on 3+ hop challenge questions (official Wikipedia paragraphs)

Source: [End-to-end Question Answering on 3+ hop challenge questions (official Wikipedia paragraphs)](<https://www.sota2.com/research/sota/end-to-end-question-answering-on-3-plus-hop-challenge-questions-official-wikipedia-paragraphs>)

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

Data updated: 2026-02-26T05:41:33.550Z

Evaluation of model performance on the 3+ hop challenge questions dataset, sourced from official Wikipedia paragraphs, using End-to-end Question Answering metrics (EM and F1).

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 | End-to-end Question Answering |
| Dataset (including any stated split) | 3+ hop challenge questions (official Wikipedia paragraphs) |
| Metric | EM |
| 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

- [EM](<https://www.sota2.com/research/sota/end-to-end-question-answering-on-3-plus-hop-challenge-questions-official-wikipedia-paragraphs.md?metric=EM>): Higher is better; unit: PERCENT.

- [F1](<https://www.sota2.com/research/sota/end-to-end-question-answering-on-3-plus-hop-challenge-questions-official-wikipedia-paragraphs.md?metric=F1>): Higher is better; unit: PERCENT.

## Recorded results

| Method | EM | Result date | Report paper |
| --- | --- | --- | --- |
| IRRR | 32.5 | Not recorded | 68fa6b2c1e209131cea2c899 |
| IRRR | 29.2 | Not recorded | 68fa6b2c1e209131cea2c899 |
| GRR | 27.2 | Not recorded | 68fa6b2c1e209131cea2c899 |

### IRRR

EM: 32.5. Result ID: 698e6537ffc3901de09d3b7e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b2c1e209131cea2c899.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b2c1e209131cea2c899.md>)

[Implementation](<https://github.com/beerqa/IRRR>)

Recorded evaluation configuration: {"Training Data":"SQUAD + HotpotQA"}. This record does not assert independent reproduction.

### IRRR

EM: 29.2. Result ID: 698e6537ffc3901de09d3b7d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b2c1e209131cea2c899.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b2c1e209131cea2c899.md>)

[Implementation](<https://github.com/beerqa/IRRR>)

Recorded evaluation configuration: {"Training Data":"HotpotQA"}. This record does not assert independent reproduction.

### GRR

EM: 27.2. Result ID: 698e6537ffc3901de09d3b7c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b2c1e209131cea2c899.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa670a1e209131cea1406e.md>)

[Implementation](<https://github.com/AkariAsai/learning_to_retrieve_reasoning_paths>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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