# Long-Form Question Answering on ELI5 (val)

Source: [Long-Form Question Answering on ELI5 (val)](<https://www.sota2.com/research/sota/long-form-question-answering-on-eli5-val>)

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

Data updated: 2026-02-27T06:45:35.927Z

Evaluation of model performance on the validation split of the ELI5 dataset for Long-Form Question Answering, measured using F1 and ROUGE-L.

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 | Long-Form Question Answering |
| Dataset (including any stated split) | ELI5 (val) |
| Metric | F1 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 11 of 11 |
| 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

- [F1](<https://www.sota2.com/research/sota/long-form-question-answering-on-eli5-val.md?metric=F1>): Higher is better; unit: PERCENT.

- [ROUGE-L](<https://www.sota2.com/research/sota/long-form-question-answering-on-eli5-val.md?metric=ROUGE-L>): Higher is better; unit: PERCENT.

## Recorded results

| Method | F1 | Result date | Report paper |
| --- | --- | --- | --- |
| best top-7 train answer | 31.5 | Not recorded | 68fa69ef1e209131cea250f1 |
| best gold answer | 29.5 | Not recorded | 68fa69ef1e209131cea250f1 |
| longest top-7 train answer | 26.9 | Not recorded | 68fa69ef1e209131cea250f1 |
| longest gold answer | 26.7 | Not recorded | 68fa69ef1e209131cea250f1 |
| best top-1 train answer | 25.9 | Not recorded | 68fa69ef1e209131cea250f1 |
| RT + C-REALM | 25.6 | Not recorded | 68fa69ef1e209131cea250f1 |
| longest top-1 train answer | 25.2 | Not recorded | 68fa69ef1e209131cea250f1 |
| BART + DPR | 18.8 | Not recorded | 68fa69ef1e209131cea250f1 |
| random train answer | 17.8 | Not recorded | 68fa69ef1e209131cea250f1 |
| RAG | 17.2 | Not recorded | 68fa69ef1e209131cea250f1 |
| copy input | 16.6 | Not recorded | 68fa69ef1e209131cea250f1 |

### best top-7 train answer

F1: 31.5. Result ID: 69a13d7435dfcabb00e5b735.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69ef1e209131cea250f1.md>)

Recorded evaluation configuration: {"Bound Type":"Upper bound"}. This record does not assert independent reproduction.

### best gold answer

F1: 29.5. Result ID: 69a13d7435dfcabb00e5b737.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69ef1e209131cea250f1.md>)

Recorded evaluation configuration: {"Bound Type":"Upper bound"}. This record does not assert independent reproduction.

### longest top-7 train answer

F1: 26.9. Result ID: 69a13d7435dfcabb00e5b732.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69ef1e209131cea250f1.md>)

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

### longest gold answer

F1: 26.7. Result ID: 69a13d7435dfcabb00e5b736.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69ef1e209131cea250f1.md>)

Recorded evaluation configuration: {"Bound Type":"Upper bound"}. This record does not assert independent reproduction.

### best top-1 train answer

F1: 25.9. Result ID: 69a13d7435dfcabb00e5b734.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69ef1e209131cea250f1.md>)

Recorded evaluation configuration: {"Bound Type":"Upper bound"}. This record does not assert independent reproduction.

### RT + C-REALM

F1: 25.6. Result ID: 69a13d7435dfcabb00e5b733.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69ef1e209131cea250f1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69ef1e209131cea250f1.md>)

[Implementation](<https://github.com/martiansideofthemoon/hurdles-longform-qa>)

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

### longest top-1 train answer

F1: 25.2. Result ID: 69a13d7435dfcabb00e5b731.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69ef1e209131cea250f1.md>)

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

### BART + DPR

F1: 18.8. Result ID: 69a13d7435dfcabb00e5b730.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69ef1e209131cea250f1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a021e209131cea255d9.md>)

[Implementation](<https://github.com/facebookresearch/KILT>)

Recorded evaluation configuration: {"Year":"2020"}. This record does not assert independent reproduction.

### random train answer

F1: 17.8. Result ID: 69a13d7435dfcabb00e5b72d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69ef1e209131cea250f1.md>)

Recorded evaluation configuration: {"Bound Type":"Lower bound"}. This record does not assert independent reproduction.

### RAG

F1: 17.2. Result ID: 69a13d7435dfcabb00e5b72f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69ef1e209131cea250f1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69b91e209131cea23c26.md>)

[Implementation](<https://github.com/huggingface/transformers/blob/master/examples/rag/>)

Recorded evaluation configuration: {"Year":"2020c"}. This record does not assert independent reproduction.

### copy input

F1: 16.6. Result ID: 69a13d7435dfcabb00e5b72e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69ef1e209131cea250f1.md>)

Recorded evaluation configuration: {"Bound Type":"Lower bound"}. This record does not assert independent reproduction.

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