# Long-form Question Answering on LFQA

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

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

Data updated: 2026-02-26T05:22:28.559Z

Evaluation of model performance on the LFQA dataset for Long-form Question Answering, measured using human/model preference scores (Fluency, Helpfulness, AIS) and inference time.

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) | LFQA |
| Metric | AIS (Decomposition) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| 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

- [AIS (Decomposition)](<https://www.sota2.com/research/sota/long-form-question-answering-on-lfqa.md?metric=AIS%20(Decomposition)>): Higher is better; unit: PERCENT.

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

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

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

- [Inference Time (sec)](<https://www.sota2.com/research/sota/long-form-question-answering-on-lfqa.md?metric=Inference%20Time%20(sec)>): Lower is better; unit: TIME.

## Recorded results

| Method | AIS (Decomposition) | Result date | Report paper |
| --- | --- | --- | --- |
| VANILLA | 90.9 | Not recorded | 68fa79001e209131cea7b93c |
| LAQuer | 59.5 | Not recorded | 68fa79001e209131cea7b93c |
| ATTR. FIRST | 53.6 | Not recorded | 68fa79001e209131cea7b93c |
| LAQuer | 50.2 | Not recorded | 68fa79001e209131cea7b93c |

### VANILLA

AIS (Decomposition): 90.9. Result ID: 6995b6123badfaffe6b69dcc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b93c.md>)

Recorded evaluation configuration: {"Attribution":"Baseline/None"}. This record does not assert independent reproduction.

### LAQuer

AIS (Decomposition): 59.5. Result ID: 6995b6123badfaffe6b69dcd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b93c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b93c.md>)

[Implementation](<https://github.com/eranhirs/LAQuer>)

Recorded evaluation configuration: {"Base Model/Method":"VANILLA","Strategy":"LLM Prompt"}. This record does not assert independent reproduction.

### ATTR. FIRST

AIS (Decomposition): 53.6. Result ID: 6995b6123badfaffe6b69dce.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b93c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74061e209131cea5f4f2.md>)

Recorded evaluation configuration: {"Attribution":"Baseline/None"}. This record does not assert independent reproduction.

### LAQuer

AIS (Decomposition): 50.2. Result ID: 6995b6123badfaffe6b69dcf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b93c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b93c.md>)

[Implementation](<https://github.com/eranhirs/LAQuer>)

Recorded evaluation configuration: {"Base Model/Method":"ATTR. FIRST","Strategy":"LLM Prompt"}. This record does not assert independent reproduction.

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