# Generative Question Answering on HaluEval (test)

Source: [Generative Question Answering on HaluEval (test)](<https://www.sota2.com/research/sota/generative-question-answering-on-halueval-test>)

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

Data updated: 2026-06-02T09:36:20.267Z

Evaluation of model performance on the test split of the HaluEval dataset for Generative Question Answering, focusing on hallucination (HALL), abstraction (ABS), factuality (FACT), and overall F-score.

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 | Generative Question Answering |
| Dataset (including any stated split) | HaluEval (test) |
| Metric | HALL Rate |
| Unit | PERCENT |
| Direction | Lower is better |
| Results loaded | 10 of 10 |
| 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

- [HALL Rate](<https://www.sota2.com/research/sota/generative-question-answering-on-halueval-test.md?metric=HALL%20Rate>): Lower is better; unit: PERCENT.

- [ABS Score](<https://www.sota2.com/research/sota/generative-question-answering-on-halueval-test.md?metric=ABS%20Score>): Higher is better; unit: PERCENT.

- [FACT Score](<https://www.sota2.com/research/sota/generative-question-answering-on-halueval-test.md?metric=FACT%20Score>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/generative-question-answering-on-halueval-test.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | HALL Rate | Result date | Report paper |
| --- | --- | --- | --- |
| Plan-and-Solve | 50.33 | Not recorded | 6a1e6393dbd7150506f85222 |
| Append-idk | 54.67 | Not recorded | 6a1e6393dbd7150506f85222 |
| RCSP | 55 | Not recorded | 6a1e6393dbd7150506f85222 |
| EmoPrompt (avg) | 55.79 | Not recorded | 6a1e6393dbd7150506f85222 |
| Plan-and-Solve | 56.67 | Not recorded | 6a1e6393dbd7150506f85222 |
| Base | 58 | Not recorded | 6a1e6393dbd7150506f85222 |
| Base | 59.67 | Not recorded | 6a1e6393dbd7150506f85222 |
| Append-idk | 60 | Not recorded | 6a1e6393dbd7150506f85222 |
| RCSP | 60.67 | Not recorded | 6a1e6393dbd7150506f85222 |
| EmoPrompt (avg) | 63.15 | Not recorded | 6a1e6393dbd7150506f85222 |

### Plan-and-Solve

HALL Rate: 50.33. Result ID: 6a1ea334fbc373592c191fa0.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e6393dbd7150506f85222.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f7f1e209131cea45fd1.md>)

Recorded evaluation configuration: {"Backbone":"Llama 3.1-8b"}. This record does not assert independent reproduction.

### Append-idk

HALL Rate: 54.67. Result ID: 6a1ea334fbc373592c191f85.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e6393dbd7150506f85222.md>)

Recorded evaluation configuration: {"Backbone":"Gemma-3-12b-pt"}. This record does not assert independent reproduction.

### RCSP

HALL Rate: 55. Result ID: 6a1ea334fbc373592c191f88.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e6393dbd7150506f85222.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a1e6393dbd7150506f85222.md>)

Recorded evaluation configuration: {"Backbone":"Gemma-3-12b-pt"}. This record does not assert independent reproduction.

### EmoPrompt (avg)

HALL Rate: 55.79. Result ID: 6a1ea334fbc373592c191f86.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e6393dbd7150506f85222.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa712b1e209131cea4fa5c.md>)

Recorded evaluation configuration: {"Backbone":"Gemma-3-12b-pt"}. This record does not assert independent reproduction.

### Plan-and-Solve

HALL Rate: 56.67. Result ID: 6a1ea334fbc373592c191f87.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e6393dbd7150506f85222.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f7f1e209131cea45fd1.md>)

Recorded evaluation configuration: {"Backbone":"Gemma-3-12b-pt"}. This record does not assert independent reproduction.

### Base

HALL Rate: 58. Result ID: 6a1ea334fbc373592c191f84.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e6393dbd7150506f85222.md>)

Recorded evaluation configuration: {"Backbone":"Gemma-3-12b-pt"}. This record does not assert independent reproduction.

### Base

HALL Rate: 59.67. Result ID: 6a1ea334fbc373592c191f9d.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e6393dbd7150506f85222.md>)

Recorded evaluation configuration: {"Backbone":"Llama 3.1-8b"}. This record does not assert independent reproduction.

### Append-idk

HALL Rate: 60. Result ID: 6a1ea334fbc373592c191f9e.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e6393dbd7150506f85222.md>)

Recorded evaluation configuration: {"Backbone":"Llama 3.1-8b"}. This record does not assert independent reproduction.

### RCSP

HALL Rate: 60.67. Result ID: 6a1ea334fbc373592c191fa1.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e6393dbd7150506f85222.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a1e6393dbd7150506f85222.md>)

Recorded evaluation configuration: {"Backbone":"Llama 3.1-8b"}. This record does not assert independent reproduction.

### EmoPrompt (avg)

HALL Rate: 63.15. Result ID: 6a1ea334fbc373592c191f9f.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e6393dbd7150506f85222.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa712b1e209131cea4fa5c.md>)

Recorded evaluation configuration: {"Backbone":"Llama 3.1-8b"}. This record does not assert independent reproduction.

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