# Pairwise Preference Comparison on Qwen2.5-3B responses (test)

Source: [Pairwise Preference Comparison on Qwen2.5-3B responses (test)](<https://www.sota2.com/research/sota/pairwise-preference-comparison-on-qwen2-5-3b-responses-test>)

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

Data updated: 2026-02-26T05:36:18.304Z

Evaluation of model performance using Pairwise Preference Comparison on the test set derived from Qwen2.5-3B responses, measured by Average preference 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 | Pairwise Preference Comparison |
| Dataset (including any stated split) | Qwen2.5-3B responses (test) |
| Metric | Avg Preference Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 30 of 30 |
| 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

- [Avg Preference Score](<https://www.sota2.com/research/sota/pairwise-preference-comparison-on-qwen2-5-3b-responses-test.md?metric=Avg%20Preference%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Avg Preference Score | Result date | Report paper |
| --- | --- | --- | --- |
| NASH-MD-PG | 82.7 | Not recorded | 6944dc1f423ea42265729bf8 |
| STACKELBERGGDA | 82.1 | Not recorded | 6944dc1f423ea42265729bf8 |
| NASH-MD-PG | 81.3 | Not recorded | 6944dc1f423ea42265729bf8 |
| STACKELBERGGDA | 81.3 | Not recorded | 6944dc1f423ea42265729bf8 |
| NASH-MD-PG | 80.1 | Not recorded | 6944dc1f423ea42265729bf8 |
| STACKELBERGGDA | 79.9 | Not recorded | 6944dc1f423ea42265729bf8 |
| STACKELBERGGDA | 69.6 | Not recorded | 6944dc1f423ea42265729bf8 |
| STACKELBERGGDA | 68.8 | Not recorded | 6944dc1f423ea42265729bf8 |
| STACKELBERGGDA | 68.1 | Not recorded | 6944dc1f423ea42265729bf8 |
| STACKELBERGGDA | 66.2 | Not recorded | 6944dc1f423ea42265729bf8 |
| NASH-MD-PG | 63.8 | Not recorded | 6944dc1f423ea42265729bf8 |
| RLOO | 60.1 | Not recorded | 6944dc1f423ea42265729bf8 |
| RLOO | 58.8 | Not recorded | 6944dc1f423ea42265729bf8 |
| NASH-MD-PG | 50.7 | Not recorded | 6944dc1f423ea42265729bf8 |
| QWEN2.5-0.5B | 50.4 | Not recorded | 6944dc1f423ea42265729bf8 |
| QWEN2.5-3B | 49.6 | Not recorded | 6944dc1f423ea42265729bf8 |
| STACKELBERGGDA | 49.3 | Not recorded | 6944dc1f423ea42265729bf8 |
| QWEN2.5-3B | 41.2 | Not recorded | 6944dc1f423ea42265729bf8 |
| QWEN2.5-0.5B | 39.9 | Not recorded | 6944dc1f423ea42265729bf8 |
| STACKELBERGGDA | 36.2 | Not recorded | 6944dc1f423ea42265729bf8 |
| RLOO | 33.8 | Not recorded | 6944dc1f423ea42265729bf8 |
| QWEN2.5-3B | 31.9 | Not recorded | 6944dc1f423ea42265729bf8 |
| STACKELBERGGDA | 31.2 | Not recorded | 6944dc1f423ea42265729bf8 |
| QWEN2.5-0.5B | 30.4 | Not recorded | 6944dc1f423ea42265729bf8 |
| RLOO | 20.1 | Not recorded | 6944dc1f423ea42265729bf8 |
| QWEN2.5-3B | 19.9 | Not recorded | 6944dc1f423ea42265729bf8 |
| QWEN2.5-0.5B | 18.7 | Not recorded | 6944dc1f423ea42265729bf8 |
| QWEN2.5-0.5B | 18.7 | Not recorded | 6944dc1f423ea42265729bf8 |
| QWEN2.5-3B | 17.9 | Not recorded | 6944dc1f423ea42265729bf8 |
| RLOO | 17.3 | Not recorded | 6944dc1f423ea42265729bf8 |

### NASH-MD-PG

Avg Preference Score: 82.7. Result ID: 699952975698665ec4bd688f.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa739a1e209131cea5d8b7.md>)

Recorded evaluation configuration: {"Comparison model":"RLOO"}. This record does not assert independent reproduction.

### STACKELBERGGDA

Avg Preference Score: 82.1. Result ID: 699952975698665ec4bd6898.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

Recorded evaluation configuration: {"Role":"Follower","Comparison model":"QWEN2.5-3B"}. This record does not assert independent reproduction.

### NASH-MD-PG

Avg Preference Score: 81.3. Result ID: 699952975698665ec4bd688d.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa739a1e209131cea5d8b7.md>)

Recorded evaluation configuration: {"Comparison model":"QWEN2.5-0.5B"}. This record does not assert independent reproduction.

### STACKELBERGGDA

Avg Preference Score: 81.3. Result ID: 699952975698665ec4bd6897.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

Recorded evaluation configuration: {"Role":"Follower","Comparison model":"QWEN2.5-0.5B"}. This record does not assert independent reproduction.

### NASH-MD-PG

Avg Preference Score: 80.1. Result ID: 699952975698665ec4bd688e.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa739a1e209131cea5d8b7.md>)

Recorded evaluation configuration: {"Comparison model":"QWEN2.5-3B"}. This record does not assert independent reproduction.

### STACKELBERGGDA

Avg Preference Score: 79.9. Result ID: 699952975698665ec4bd6899.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

Recorded evaluation configuration: {"Role":"Follower","Comparison model":"RLOO"}. This record does not assert independent reproduction.

### STACKELBERGGDA

Avg Preference Score: 69.6. Result ID: 699952975698665ec4bd6892.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

Recorded evaluation configuration: {"Role":"Leader","Comparison model":"QWEN2.5-0.5B"}. This record does not assert independent reproduction.

### STACKELBERGGDA

Avg Preference Score: 68.8. Result ID: 699952975698665ec4bd689b.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

Recorded evaluation configuration: {"Role":"Follower","Comparison model":"STACKELBERGGDA LEADER"}. This record does not assert independent reproduction.

### STACKELBERGGDA

Avg Preference Score: 68.1. Result ID: 699952975698665ec4bd6893.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

Recorded evaluation configuration: {"Role":"Leader","Comparison model":"QWEN2.5-3B"}. This record does not assert independent reproduction.

### STACKELBERGGDA

Avg Preference Score: 66.2. Result ID: 699952975698665ec4bd6894.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

Recorded evaluation configuration: {"Role":"Leader","Comparison model":"RLOO"}. This record does not assert independent reproduction.

### NASH-MD-PG

Avg Preference Score: 63.8. Result ID: 699952975698665ec4bd6890.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa739a1e209131cea5d8b7.md>)

Recorded evaluation configuration: {"Comparison model":"STACKELBERGGDA LEADER"}. This record does not assert independent reproduction.

### RLOO

Avg Preference Score: 60.1. Result ID: 699952975698665ec4bd6888.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa725a1e209131cea55df7.md>)

Recorded evaluation configuration: {"Comparison model":"QWEN2.5-0.5B"}. This record does not assert independent reproduction.

### RLOO

Avg Preference Score: 58.8. Result ID: 699952975698665ec4bd6889.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa725a1e209131cea55df7.md>)

Recorded evaluation configuration: {"Comparison model":"QWEN2.5-3B"}. This record does not assert independent reproduction.

### NASH-MD-PG

Avg Preference Score: 50.7. Result ID: 699952975698665ec4bd6891.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa739a1e209131cea5d8b7.md>)

Recorded evaluation configuration: {"Comparison model":"STACKELBERGGDA FOLLOWER"}. This record does not assert independent reproduction.

### QWEN2.5-0.5B

Avg Preference Score: 50.4. Result ID: 699952975698665ec4bd687e.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Comparison model":"QWEN2.5-3B"}. This record does not assert independent reproduction.

### QWEN2.5-3B

Avg Preference Score: 49.6. Result ID: 699952975698665ec4bd6883.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Comparison model":"QWEN2.5-0.5B"}. This record does not assert independent reproduction.

### STACKELBERGGDA

Avg Preference Score: 49.3. Result ID: 699952975698665ec4bd689a.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

Recorded evaluation configuration: {"Role":"Follower","Comparison model":"NASH-MD-PG"}. This record does not assert independent reproduction.

### QWEN2.5-3B

Avg Preference Score: 41.2. Result ID: 699952975698665ec4bd6884.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Comparison model":"RLOO"}. This record does not assert independent reproduction.

### QWEN2.5-0.5B

Avg Preference Score: 39.9. Result ID: 699952975698665ec4bd687f.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Comparison model":"RLOO"}. This record does not assert independent reproduction.

### STACKELBERGGDA

Avg Preference Score: 36.2. Result ID: 699952975698665ec4bd6895.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

Recorded evaluation configuration: {"Role":"Leader","Comparison model":"NASH-MD-PG"}. This record does not assert independent reproduction.

### RLOO

Avg Preference Score: 33.8. Result ID: 699952975698665ec4bd688b.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa725a1e209131cea55df7.md>)

Recorded evaluation configuration: {"Comparison model":"STACKELBERGGDA LEADER"}. This record does not assert independent reproduction.

### QWEN2.5-3B

Avg Preference Score: 31.9. Result ID: 699952975698665ec4bd6886.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Comparison model":"STACKELBERGGDA LEADER"}. This record does not assert independent reproduction.

### STACKELBERGGDA

Avg Preference Score: 31.2. Result ID: 699952975698665ec4bd6896.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

Recorded evaluation configuration: {"Role":"Leader","Comparison model":"STACKELBERGGDA FOLLOWER"}. This record does not assert independent reproduction.

### QWEN2.5-0.5B

Avg Preference Score: 30.4. Result ID: 699952975698665ec4bd6881.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Comparison model":"STACKELBERGGDA LEADER"}. This record does not assert independent reproduction.

### RLOO

Avg Preference Score: 20.1. Result ID: 699952975698665ec4bd688c.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa725a1e209131cea55df7.md>)

Recorded evaluation configuration: {"Comparison model":"STACKELBERGGDA FOLLOWER"}. This record does not assert independent reproduction.

### QWEN2.5-3B

Avg Preference Score: 19.9. Result ID: 699952975698665ec4bd6885.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Comparison model":"NASH-MD-PG"}. This record does not assert independent reproduction.

### QWEN2.5-0.5B

Avg Preference Score: 18.7. Result ID: 699952975698665ec4bd6880.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Comparison model":"NASH-MD-PG"}. This record does not assert independent reproduction.

### QWEN2.5-0.5B

Avg Preference Score: 18.7. Result ID: 699952975698665ec4bd6882.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Comparison model":"STACKELBERGGDA FOLLOWER"}. This record does not assert independent reproduction.

### QWEN2.5-3B

Avg Preference Score: 17.9. Result ID: 699952975698665ec4bd6887.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Comparison model":"STACKELBERGGDA FOLLOWER"}. This record does not assert independent reproduction.

### RLOO

Avg Preference Score: 17.3. Result ID: 699952975698665ec4bd688a.

[Reporting paper](<https://www.sota2.com/research/paper/6944dc1f423ea42265729bf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa725a1e209131cea55df7.md>)

Recorded evaluation configuration: {"Comparison model":"NASH-MD-PG"}. This record does not assert independent reproduction.

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