# Reward Performance on Twenty Questions High-data

Source: [Reward Performance on Twenty Questions High-data](<https://www.sota2.com/research/sota/reward-performance-on-twenty-questions-high-data>)

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

Data updated: 2026-05-26T06:47:51.832Z

Evaluation of model performance on the Twenty Questions High-data dataset, specifically measuring test-time reward.

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 | Reward Performance |
| Dataset (including any stated split) | Twenty Questions High-data |
| Metric | Test-Time Reward |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [Test-Time Reward](<https://www.sota2.com/research/sota/reward-performance-on-twenty-questions-high-data.md?metric=Test-Time%20Reward>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Test-Time Reward | Result date | Report paper |
| --- | --- | --- | --- |
| ILQL | 5.83 | Not recorded | 6a15293bdbd7150506f7e0c4 |
| ILQL | 5.38 | Not recorded | 6a15293bdbd7150506f7e0c4 |
| MC | 3.83 | Not recorded | 6a15293bdbd7150506f7e0c4 |
| MC | 3.48 | Not recorded | 6a15293bdbd7150506f7e0c4 |
| ILQL | 2.6 | Not recorded | 6a15293bdbd7150506f7e0c4 |
| MC | 2.22 | Not recorded | 6a15293bdbd7150506f7e0c4 |
| MC | 2.19 | Not recorded | 6a15293bdbd7150506f7e0c4 |
| ILQL | 2.04 | Not recorded | 6a15293bdbd7150506f7e0c4 |

### ILQL

Test-Time Reward: 5.83. Result ID: 6a15416efbc373592c0c01a6.

[Reporting paper](<https://www.sota2.com/research/paper/6a15293bdbd7150506f7e0c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a15293bdbd7150506f7e0c4.md>)

Recorded evaluation configuration: {"training\_regime":"BOOST"}. This record does not assert independent reproduction.

### ILQL

Test-Time Reward: 5.38. Result ID: 6a15416efbc373592c0c01a5.

[Reporting paper](<https://www.sota2.com/research/paper/6a15293bdbd7150506f7e0c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a15293bdbd7150506f7e0c4.md>)

Recorded evaluation configuration: {"training\_regime":"Bilevel Only"}. This record does not assert independent reproduction.

### MC

Test-Time Reward: 3.83. Result ID: 6a15416efbc373592c0c01a1.

[Reporting paper](<https://www.sota2.com/research/paper/6a15293bdbd7150506f7e0c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a15293bdbd7150506f7e0c4.md>)

Recorded evaluation configuration: {"training\_regime":"Bilevel Only"}. This record does not assert independent reproduction.

### MC

Test-Time Reward: 3.48. Result ID: 6a15416efbc373592c0c01a2.

[Reporting paper](<https://www.sota2.com/research/paper/6a15293bdbd7150506f7e0c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a15293bdbd7150506f7e0c4.md>)

Recorded evaluation configuration: {"training\_regime":"BOOST"}. This record does not assert independent reproduction.

### ILQL

Test-Time Reward: 2.6. Result ID: 6a15416efbc373592c0c01a3.

[Reporting paper](<https://www.sota2.com/research/paper/6a15293bdbd7150506f7e0c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a15293bdbd7150506f7e0c4.md>)

Recorded evaluation configuration: {"training\_regime":"Vanilla"}. This record does not assert independent reproduction.

### MC

Test-Time Reward: 2.22. Result ID: 6a15416efbc373592c0c019f.

[Reporting paper](<https://www.sota2.com/research/paper/6a15293bdbd7150506f7e0c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a15293bdbd7150506f7e0c4.md>)

Recorded evaluation configuration: {"training\_regime":"Vanilla"}. This record does not assert independent reproduction.

### MC

Test-Time Reward: 2.19. Result ID: 6a15416efbc373592c0c01a0.

[Reporting paper](<https://www.sota2.com/research/paper/6a15293bdbd7150506f7e0c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a15293bdbd7150506f7e0c4.md>)

Recorded evaluation configuration: {"training\_regime":"Vanilla+Syn"}. This record does not assert independent reproduction.

### ILQL

Test-Time Reward: 2.04. Result ID: 6a15416efbc373592c0c01a4.

[Reporting paper](<https://www.sota2.com/research/paper/6a15293bdbd7150506f7e0c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a15293bdbd7150506f7e0c4.md>)

Recorded evaluation configuration: {"training\_regime":"Vanilla+Syn"}. This record does not assert independent reproduction.

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