# Explanation Faithfulness Evaluation on Math Book (test)

Source: [Explanation Faithfulness Evaluation on Math Book (test)](<https://www.sota2.com/research/sota/explanation-faithfulness-evaluation-on-math-book-test>)

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

Data updated: 2026-07-08T07:41:24.736Z

Evaluation of explanation faithfulness on the test split of the Math Book dataset, measured using Greedy and Maj@16 scores.

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 | Explanation Faithfulness Evaluation |
| Dataset (including any stated split) | Math Book (test) |
| Metric | Greedy Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 7 of 7 |
| 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

- [Greedy Score](<https://www.sota2.com/research/sota/explanation-faithfulness-evaluation-on-math-book-test.md?metric=Greedy%20Score>): Higher is better; unit: PERCENT.

- [Majority Voting Score (@16)](<https://www.sota2.com/research/sota/explanation-faithfulness-evaluation-on-math-book-test.md?metric=Majority%20Voting%20Score%20(%4016)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Greedy Score | Result date | Report paper |
| --- | --- | --- | --- |
| DPO + RMD | 31.4 | Not recorded | 68fa7a0d1e209131cea80d2f |
| DPO + RM | 30.8 | Not recorded | 68fa7a0d1e209131cea80d2f |
| DPO + RMD | 28.9 | Not recorded | 68fa7a0d1e209131cea80d2f |
| DPO + RM | 27.8 | Not recorded | 68fa7a0d1e209131cea80d2f |
| DPO + RMC | 27.7 | Not recorded | 68fa7a0d1e209131cea80d2f |
| DPO + RMC | 25.2 | Not recorded | 68fa7a0d1e209131cea80d2f |
| Base | 23.6 | Not recorded | 68fa7a0d1e209131cea80d2f |

### DPO + RMD

Greedy Score: 31.4. Result ID: 6a4dff1460b88bb339d75b98.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a0d1e209131cea80d2f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a0d1e209131cea80d2f.md>)

Recorded evaluation configuration: {"Base Model":"LLAMA-3.2-3B-IT","Training Method":"DPO","Reward Model":"SK-GEMMA-27B"}. This record does not assert independent reproduction.

### DPO + RM

Greedy Score: 30.8. Result ID: 6a4dff1460b88bb339d75b97.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a0d1e209131cea80d2f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74511e209131cea6105d.md>)

Recorded evaluation configuration: {"Base Model":"LLAMA-3.2-3B-IT","Training Method":"DPO","Reward Model":"SK-GEMMA-27B"}. This record does not assert independent reproduction.

### DPO + RMD

Greedy Score: 28.9. Result ID: 6a4dff1460b88bb339d75b95.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a0d1e209131cea80d2f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a0d1e209131cea80d2f.md>)

Recorded evaluation configuration: {"Base Model":"LLAMA-3.2-3B-IT","Training Method":"DPO","Reward Model":"SK-LLAMA-8B"}. This record does not assert independent reproduction.

### DPO + RM

Greedy Score: 27.8. Result ID: 6a4dff1460b88bb339d75b94.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a0d1e209131cea80d2f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74511e209131cea6105d.md>)

Recorded evaluation configuration: {"Base Model":"LLAMA-3.2-3B-IT","Training Method":"DPO","Reward Model":"SK-LLAMA-8B"}. This record does not assert independent reproduction.

### DPO + RMC

Greedy Score: 27.7. Result ID: 6a4dff1460b88bb339d75b99.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a0d1e209131cea80d2f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a0d1e209131cea80d2f.md>)

Recorded evaluation configuration: {"Base Model":"LLAMA-3.2-3B-IT","Training Method":"DPO","Reward Model":"SK-GEMMA-27B"}. This record does not assert independent reproduction.

### DPO + RMC

Greedy Score: 25.2. Result ID: 6a4dff1460b88bb339d75b96.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a0d1e209131cea80d2f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a0d1e209131cea80d2f.md>)

Recorded evaluation configuration: {"Base Model":"LLAMA-3.2-3B-IT","Training Method":"DPO","Reward Model":"SK-LLAMA-8B"}. This record does not assert independent reproduction.

### Base

Greedy Score: 23.6. Result ID: 6a4dff1460b88bb339d75b93.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a0d1e209131cea80d2f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Base Model":"LLAMA-3.2-3B-IT","Training Method":"Base","Reward Model":"None"}. This record does not assert independent reproduction.

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