# End-to-end Question Answering on AGORA (test)

Source: [End-to-end Question Answering on AGORA (test)](<https://www.sota2.com/research/sota/end-to-end-question-answering-on-agora-test>)

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

Data updated: 2026-03-26T15:56:04.861Z

Evaluation of model performance on the test split of the AGORA dataset for End-to-end Question Answering tasks, reporting metrics such as Relevancy and 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 | End-to-end Question Answering |
| Dataset (including any stated split) | AGORA (test) |
| Metric | Relevancy |
| 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

- [Relevancy](<https://www.sota2.com/research/sota/end-to-end-question-answering-on-agora-test.md?metric=Relevancy>): Higher is better; unit: PERCENT.

- [Accuracy](<https://www.sota2.com/research/sota/end-to-end-question-answering-on-agora-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Relevancy | Result date | Report paper |
| --- | --- | --- | --- |
| Base ColBERT / DPO Mistral | 74.7 | Not recorded | 69c4bd9d0636e5bf4afc768b |
| Base ColBERT / Base Mistral | 74.6 | Not recorded | 69c4bd9d0636e5bf4afc768b |
| CL ColBERT / Base Mistral | 74.4 | Not recorded | 69c4bd9d0636e5bf4afc768b |
| GPT-5.4 | 74.4 | Not recorded | 69c4bd9d0636e5bf4afc768b |
| Base Mistral (w/o RAG) | 73.3 | Not recorded | 69c4bd9d0636e5bf4afc768b |
| CL ColBERT / DPO Mistral | 70 | Not recorded | 69c4bd9d0636e5bf4afc768b |

### Base ColBERT / DPO Mistral

Relevancy: 74.7. Result ID: 69c5570547f34ce924afb493.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bd9d0636e5bf4afc768b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c4bd9d0636e5bf4afc768b.md>)

Recorded evaluation configuration: {"Retriever":"Base ColBERTv2","Generator":"Mistral-7B-Instruct aligned with DPO"}. This record does not assert independent reproduction.

### Base ColBERT / Base Mistral

Relevancy: 74.6. Result ID: 69c5570547f34ce924afb491.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bd9d0636e5bf4afc768b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cf41e209131cea36cce.md>)

[Implementation](<https://github.com/stanford-futuredata/ColBERT>)

Recorded evaluation configuration: {"Retriever":"Base ColBERTv2","Generator":"Base Mistral-7B-Instruct"}. This record does not assert independent reproduction.

### CL ColBERT / Base Mistral

Relevancy: 74.4. Result ID: 69c5570547f34ce924afb492.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bd9d0636e5bf4afc768b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c4bd9d0636e5bf4afc768b.md>)

Recorded evaluation configuration: {"Retriever":"Fine-tuned ColBERTv2 (Contrastive Learning)","Generator":"Base Mistral-7B-Instruct"}. This record does not assert independent reproduction.

### GPT-5.4

Relevancy: 74.4. Result ID: 69c5570547f34ce924afb495.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bd9d0636e5bf4afc768b.md>)

Recorded evaluation configuration: {"model\_family":"GPT-5.4"}. This record does not assert independent reproduction.

### Base Mistral (w/o RAG)

Relevancy: 73.3. Result ID: 69c5570547f34ce924afb490.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bd9d0636e5bf4afc768b.md>)

Recorded evaluation configuration: {"RAG":"false","Generator":"Mistral-7B-Instruct"}. This record does not assert independent reproduction.

### CL ColBERT / DPO Mistral

Relevancy: 70. Result ID: 69c5570547f34ce924afb494.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bd9d0636e5bf4afc768b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c4bd9d0636e5bf4afc768b.md>)

Recorded evaluation configuration: {"Retriever":"Fine-tuned ColBERTv2 (Contrastive Learning)","Generator":"Mistral-7B-Instruct aligned with DPO"}. This record does not assert independent reproduction.

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