# Question Answering on TriviaQA (Positional Accuracy Metrics)

Source: [Question Answering on TriviaQA (Positional Accuracy Metrics)](<https://www.sota2.com/research/sota/question-answering-on-triviaqa-positional-accuracy-metrics>)

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

Data updated: 2026-02-26T05:34:10.437Z

Evaluation of Question Answering performance on TriviaQA, focusing on positional accuracy metrics (Accuracy @Pos 1, @Pos 10) and benign 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 | Question Answering |
| Dataset (including any stated split) | TriviaQA |
| Metric | Accuracy (Benign) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 15 of 15 |
| 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

- [Accuracy (Benign)](<https://www.sota2.com/research/sota/question-answering-on-triviaqa-positional-accuracy-metrics.md?metric=Accuracy%20(Benign)>): Higher is better; unit: PERCENT.

- [Accuracy (@Pos 1)](<https://www.sota2.com/research/sota/question-answering-on-triviaqa-positional-accuracy-metrics.md?metric=Accuracy%20(%40Pos%201)>): Higher is better; unit: PERCENT.

- [Accuracy (@Pos 10)](<https://www.sota2.com/research/sota/question-answering-on-triviaqa-positional-accuracy-metrics.md?metric=Accuracy%20(%40Pos%2010)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (Benign) | Result date | Report paper |
| --- | --- | --- | --- |
| MIS/Sampling + MIS | 73.4 | Not recorded | 68fa7bb31e209131cea89fbe |
| Vanilla RAG | 72.8 | Not recorded | 68fa7bb31e209131cea89fbe |
| InstructRAG | 72.2 | Not recorded | 68fa7bb31e209131cea89fbe |
| AstuteRAG | 70.2 | Not recorded | 68fa7bb31e209131cea89fbe |
| InstructRAG | 70 | Not recorded | 68fa7bb31e209131cea89fbe |
| InstructRAG | 69 | Not recorded | 68fa7bb31e209131cea89fbe |
| Vanilla RAG | 68.6 | Not recorded | 68fa7bb31e209131cea89fbe |
| AstuteRAG | 68.2 | Not recorded | 68fa7bb31e209131cea89fbe |
| RobustRAG | 67.8 | Not recorded | 68fa7bb31e209131cea89fbe |
| MIS/Sampling + MIS | 65.4 | Not recorded | 68fa7bb31e209131cea89fbe |
| Vanilla RAG | 65.2 | Not recorded | 68fa7bb31e209131cea89fbe |
| MIS/Sampling + MIS | 65 | Not recorded | 68fa7bb31e209131cea89fbe |
| AstuteRAG | 61.8 | Not recorded | 68fa7bb31e209131cea89fbe |
| RobustRAG | 60.8 | Not recorded | 68fa7bb31e209131cea89fbe |
| RobustRAG | 60.8 | Not recorded | 68fa7bb31e209131cea89fbe |

### MIS/Sampling + MIS

Accuracy (Benign): 73.4. Result ID: 699a88f55698665ec41317d4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb31e209131cea89fbe.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bb31e209131cea89fbe.md>)

Recorded evaluation configuration: {"Model":"GPT-4o-mini","Number of retrieved documents (k)":10}. This record does not assert independent reproduction.

### Vanilla RAG

Accuracy (Benign): 72.8. Result ID: 699a88f55698665ec41317d0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb31e209131cea89fbe.md>)

Recorded evaluation configuration: {"Model":"GPT-4o-mini","Number of retrieved documents (k)":10}. This record does not assert independent reproduction.

### InstructRAG

Accuracy (Benign): 72.2. Result ID: 699a88f55698665ec41317c8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb31e209131cea89fbe.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa773e1e209131cea71ed6.md>)

Recorded evaluation configuration: {"Model":"Mistral-7B","Number of retrieved documents (k)":10}. This record does not assert independent reproduction.

### AstuteRAG

Accuracy (Benign): 70.2. Result ID: 699a88f55698665ec41317d1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb31e209131cea89fbe.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79061e209131cea7bc2c.md>)

Recorded evaluation configuration: {"Model":"GPT-4o-mini","Number of retrieved documents (k)":10}. This record does not assert independent reproduction.

### InstructRAG

Accuracy (Benign): 70. Result ID: 699a88f55698665ec41317cd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb31e209131cea89fbe.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa773e1e209131cea71ed6.md>)

Recorded evaluation configuration: {"Model":"Llama3.2-3B","Number of retrieved documents (k)":10}. This record does not assert independent reproduction.

### InstructRAG

Accuracy (Benign): 69. Result ID: 699a88f55698665ec41317d2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb31e209131cea89fbe.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa773e1e209131cea71ed6.md>)

Recorded evaluation configuration: {"Model":"GPT-4o-mini","Number of retrieved documents (k)":10}. This record does not assert independent reproduction.

### Vanilla RAG

Accuracy (Benign): 68.6. Result ID: 699a88f55698665ec41317c6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb31e209131cea89fbe.md>)

Recorded evaluation configuration: {"Model":"Mistral-7B","Number of retrieved documents (k)":10}. This record does not assert independent reproduction.

### AstuteRAG

Accuracy (Benign): 68.2. Result ID: 699a88f55698665ec41317cc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb31e209131cea89fbe.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79061e209131cea7bc2c.md>)

Recorded evaluation configuration: {"Model":"Llama3.2-3B","Number of retrieved documents (k)":10}. This record does not assert independent reproduction.

### RobustRAG

Accuracy (Benign): 67.8. Result ID: 699a88f55698665ec41317d3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb31e209131cea89fbe.md>)

Recorded evaluation configuration: {"Model":"GPT-4o-mini","Number of retrieved documents (k)":10}. This record does not assert independent reproduction.

### MIS/Sampling + MIS

Accuracy (Benign): 65.4. Result ID: 699a88f55698665ec41317ca.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb31e209131cea89fbe.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bb31e209131cea89fbe.md>)

Recorded evaluation configuration: {"Model":"Mistral-7B","Number of retrieved documents (k)":10}. This record does not assert independent reproduction.

### Vanilla RAG

Accuracy (Benign): 65.2. Result ID: 699a88f55698665ec41317cb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb31e209131cea89fbe.md>)

Recorded evaluation configuration: {"Model":"Llama3.2-3B","Number of retrieved documents (k)":10}. This record does not assert independent reproduction.

### MIS/Sampling + MIS

Accuracy (Benign): 65. Result ID: 699a88f55698665ec41317cf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb31e209131cea89fbe.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bb31e209131cea89fbe.md>)

Recorded evaluation configuration: {"Model":"Llama3.2-3B","Number of retrieved documents (k)":10}. This record does not assert independent reproduction.

### AstuteRAG

Accuracy (Benign): 61.8. Result ID: 699a88f55698665ec41317c7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb31e209131cea89fbe.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79061e209131cea7bc2c.md>)

Recorded evaluation configuration: {"Model":"Mistral-7B","Number of retrieved documents (k)":10}. This record does not assert independent reproduction.

### RobustRAG

Accuracy (Benign): 60.8. Result ID: 699a88f55698665ec41317c9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb31e209131cea89fbe.md>)

Recorded evaluation configuration: {"Model":"Mistral-7B","Number of retrieved documents (k)":10}. This record does not assert independent reproduction.

### RobustRAG

Accuracy (Benign): 60.8. Result ID: 699a88f55698665ec41317ce.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb31e209131cea89fbe.md>)

Recorded evaluation configuration: {"Model":"Llama3.2-3B","Number of retrieved documents (k)":10}. This record does not assert independent reproduction.

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