# Role-playing Dialogue Evaluation on FURINA-Bench English

Source: [Role-playing Dialogue Evaluation on FURINA-Bench English](<https://www.sota2.com/research/sota/role-playing-dialogue-evaluation-on-furina-bench-english>)

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

Data updated: 2026-04-07T09:32:38.562Z

Evaluation of model role-playing dialogue capabilities using the FURINA-Bench English dataset, assessing performance across dimensions including Context Reliance, Factual Recall, Reflective Reasoning, Conversational Ability, and Preference Alignment.

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 | Role-playing Dialogue Evaluation |
| Dataset (including any stated split) | FURINA-Bench English |
| Metric | Context Reliance |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 15 of 15 |
| 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

- [Context Reliance](<https://www.sota2.com/research/sota/role-playing-dialogue-evaluation-on-furina-bench-english.md?metric=Context%20Reliance>): Higher is better; unit: SCALAR.

- [Factual Recall](<https://www.sota2.com/research/sota/role-playing-dialogue-evaluation-on-furina-bench-english.md?metric=Factual%20Recall>): Higher is better; unit: SCALAR.

- [Reflective Reasoning](<https://www.sota2.com/research/sota/role-playing-dialogue-evaluation-on-furina-bench-english.md?metric=Reflective%20Reasoning>): Higher is better; unit: SCALAR.

- [Conversational Ability](<https://www.sota2.com/research/sota/role-playing-dialogue-evaluation-on-furina-bench-english.md?metric=Conversational%20Ability>): Higher is better; unit: SCALAR.

- [Preference Alignment](<https://www.sota2.com/research/sota/role-playing-dialogue-evaluation-on-furina-bench-english.md?metric=Preference%20Alignment>): Higher is better; unit: SCALAR.

- [Weighted Average Score](<https://www.sota2.com/research/sota/role-playing-dialogue-evaluation-on-furina-bench-english.md?metric=Weighted%20Average%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Context Reliance | Result date | Report paper |
| --- | --- | --- | --- |
| o3 | 42.99 | Not recorded | 68fa7c101e209131cea8c38e |
| Deepseek-R1 | 36.31 | Not recorded | 68fa7c101e209131cea8c38e |
| Claude-4-Sonnet | 32.62 | Not recorded | 68fa7c101e209131cea8c38e |
| Claude-4-Sonnet | 30.8 | Not recorded | 68fa7c101e209131cea8c38e |
| Qwen3-32B | 28.49 | Not recorded | 68fa7c101e209131cea8c38e |
| Qwen3-235B | 27.37 | Not recorded | 68fa7c101e209131cea8c38e |
| GPT-4o | 25.7 | Not recorded | 68fa7c101e209131cea8c38e |
| Deepseek-V3 | 22.51 | Not recorded | 68fa7c101e209131cea8c38e |
| Qwen3-235B | 19.61 | Not recorded | 68fa7c101e209131cea8c38e |
| Qwen3-32B | 15.98 | Not recorded | 68fa7c101e209131cea8c38e |
| Llama3.1-70B | 14.62 | Not recorded | 68fa7c101e209131cea8c38e |
| Qwen3-8B | 14.28 | Not recorded | 68fa7c101e209131cea8c38e |
| Qwen3-8B | 12.78 | Not recorded | 68fa7c101e209131cea8c38e |
| CoSER-70B | 11.16 | Not recorded | 68fa7c101e209131cea8c38e |
| Llama3.1-8B | 10.85 | Not recorded | 68fa7c101e209131cea8c38e |

### o3

Context Reliance: 42.99. Result ID: 69d4cf3547f34ce924bceeb2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c38e.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Deepseek-R1

Context Reliance: 36.31. Result ID: 69d4cf3547f34ce924bceeaf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c38e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76a91e209131cea6e3eb.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Claude-4-Sonnet

Context Reliance: 32.62. Result ID: 69d4cf3547f34ce924bceeb0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c38e.md>)

Recorded evaluation configuration: {"thinking":"false"}. This record does not assert independent reproduction.

### Claude-4-Sonnet

Context Reliance: 30.8. Result ID: 69d4cf3547f34ce924bceeb1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c38e.md>)

Recorded evaluation configuration: {"thinking":"true"}. This record does not assert independent reproduction.

### Qwen3-32B

Context Reliance: 28.49. Result ID: 69d4cf3547f34ce924bceea8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c38e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"thinking":"true"}. This record does not assert independent reproduction.

### Qwen3-235B

Context Reliance: 27.37. Result ID: 69d4cf3547f34ce924bceeac.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c38e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"thinking":"true"}. This record does not assert independent reproduction.

### GPT-4o

Context Reliance: 25.7. Result ID: 69d4cf3547f34ce924bceead.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c38e.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Deepseek-V3

Context Reliance: 22.51. Result ID: 69d4cf3547f34ce924bceeae.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c38e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77001e209131cea7066a.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Qwen3-235B

Context Reliance: 19.61. Result ID: 69d4cf3547f34ce924bceeab.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c38e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"thinking":"false"}. This record does not assert independent reproduction.

### Qwen3-32B

Context Reliance: 15.98. Result ID: 69d4cf3547f34ce924bceea7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c38e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"thinking":"false"}. This record does not assert independent reproduction.

### Llama3.1-70B

Context Reliance: 14.62. Result ID: 69d4cf3547f34ce924bceea9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c38e.md>)

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

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

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Qwen3-8B

Context Reliance: 14.28. Result ID: 69d4cf3547f34ce924bceea6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c38e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"thinking":"true"}. This record does not assert independent reproduction.

### Qwen3-8B

Context Reliance: 12.78. Result ID: 69d4cf3547f34ce924bceea5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c38e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"thinking":"false"}. This record does not assert independent reproduction.

### CoSER-70B

Context Reliance: 11.16. Result ID: 69d4cf3547f34ce924bceeaa.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c38e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78d31e209131cea7a9fb.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Llama3.1-8B

Context Reliance: 10.85. Result ID: 69d4cf3547f34ce924bceea4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c38e.md>)

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

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

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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