# Operational decision-making on hand-crafted simulator

Source: [Operational decision-making on hand-crafted simulator](<https://www.sota2.com/research/sota/operational-decision-making-on-hand-crafted-simulator>)

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

Data updated: 2026-03-27T12:31:09.438Z

Evaluation of model performance on Operational decision-making using the hand-crafted simulator dataset, measured by Throughput Improvement and 95% CI.

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 | Operational decision-making |
| Dataset (including any stated split) | hand-crafted simulator |
| Metric | Throughput Improvement |
| 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

- [Throughput Improvement](<https://www.sota2.com/research/sota/operational-decision-making-on-hand-crafted-simulator.md?metric=Throughput%20Improvement>): Higher is better; unit: PERCENT.

- [95% CI](<https://www.sota2.com/research/sota/operational-decision-making-on-hand-crafted-simulator.md?metric=95%25%20CI>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Throughput Improvement | Result date | Report paper |
| --- | --- | --- | --- |
| SFT + DPO | 0.6 | Not recorded | 69c60ee90636e5bf4afc83b6 |
| SFT + DPO | 0.3 | Not recorded | 69c60ee90636e5bf4afc83b6 |
| DPO only | 0 | Not recorded | 69c60ee90636e5bf4afc83b6 |
| Human Decisions | 0 | Not recorded | 69c60ee90636e5bf4afc83b6 |
| DPO only | -0.4 | Not recorded | 69c60ee90636e5bf4afc83b6 |
| SFT only | -0.5 | Not recorded | 69c60ee90636e5bf4afc83b6 |
| Self-Refine | -2.1 | Not recorded | 69c60ee90636e5bf4afc83b6 |

### SFT + DPO

Throughput Improvement: 0.6. Result ID: 69c6787d47f34ce924b04a51.

[Reporting paper](<https://www.sota2.com/research/paper/69c60ee90636e5bf4afc83b6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c60ee90636e5bf4afc83b6.md>)

Recorded evaluation configuration: {"Training iterations":"2","Base Model":"Qwen2.5-14B-Instruct"}. This record does not assert independent reproduction.

### SFT + DPO

Throughput Improvement: 0.3. Result ID: 69c6787d47f34ce924b04a52.

[Reporting paper](<https://www.sota2.com/research/paper/69c60ee90636e5bf4afc83b6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c60ee90636e5bf4afc83b6.md>)

Recorded evaluation configuration: {"Training iterations":"1","Base Model":"Qwen2.5-14B-Instruct"}. This record does not assert independent reproduction.

### DPO only

Throughput Improvement: 0. Result ID: 69c6787d47f34ce924b04a53.

[Reporting paper](<https://www.sota2.com/research/paper/69c60ee90636e5bf4afc83b6.md>)

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

Recorded evaluation configuration: {"Training iterations":"2","Base Model":"Qwen2.5-14B-Instruct"}. This record does not assert independent reproduction.

### Human Decisions

Throughput Improvement: 0. Result ID: 69c6787d47f34ce924b04a57.

[Reporting paper](<https://www.sota2.com/research/paper/69c60ee90636e5bf4afc83b6.md>)

Recorded evaluation configuration: {"Protocol":"replayed"}. This record does not assert independent reproduction.

### DPO only

Throughput Improvement: -0.4. Result ID: 69c6787d47f34ce924b04a54.

[Reporting paper](<https://www.sota2.com/research/paper/69c60ee90636e5bf4afc83b6.md>)

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

Recorded evaluation configuration: {"Training iterations":"1","Base Model":"Qwen2.5-14B-Instruct"}. This record does not assert independent reproduction.

### SFT only

Throughput Improvement: -0.5. Result ID: 69c6787d47f34ce924b04a55.

[Reporting paper](<https://www.sota2.com/research/paper/69c60ee90636e5bf4afc83b6.md>)

Recorded evaluation configuration: {"Base Model":"Qwen2.5-14B-Instruct"}. This record does not assert independent reproduction.

### Self-Refine

Throughput Improvement: -2.1. Result ID: 69c6787d47f34ce924b04a56.

[Reporting paper](<https://www.sota2.com/research/paper/69c60ee90636e5bf4afc83b6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f7f1e209131cea45fc4.md>)

Recorded evaluation configuration: {"Base Model":"Qwen 32B"}. This record does not assert independent reproduction.

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