# Sarcasm detection on CSC (test)

Source: [Sarcasm detection on CSC (test)](<https://www.sota2.com/research/sota/sarcasm-detection-on-csc-test>)

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

Data updated: 2026-05-12T15:42:56.030Z

Evaluation of Sarcasm detection performance using the test split of the CSC dataset.

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 | Sarcasm detection |
| Dataset (including any stated split) | CSC (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 14 of 14 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/sarcasm-detection-on-csc-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/sarcasm-detection-on-csc-test.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

- [Mean Absolute Error](<https://www.sota2.com/research/sota/sarcasm-detection-on-csc-test.md?metric=Mean%20Absolute%20Error>): Lower is better; unit: SCALAR.

- [Correlation Coefficient](<https://www.sota2.com/research/sota/sarcasm-detection-on-csc-test.md?metric=Correlation%20Coefficient>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| DPUA | 85.57 | Not recorded | 6a02b47adbc836b87d824c13 |
| DPUA | 85.57 | Not recorded | 6a02b47adbc836b87d824c13 |
| w/ SFT | 85.12 | Not recorded | 6a02b47adbc836b87d824c13 |
| w/ GRPO | 85.12 | Not recorded | 6a02b47adbc836b87d824c13 |
| w/ GRPO\* | 84.97 | Not recorded | 6a02b47adbc836b87d824c13 |
| w/ GRPO | 84.59 | Not recorded | 6a02b47adbc836b87d824c13 |
| w/ SFT | 84.38 | Not recorded | 6a02b47adbc836b87d824c13 |
| w/ GRPO\* | 84.08 | Not recorded | 6a02b47adbc836b87d824c13 |
| w/ SFT\* | 84.04 | Not recorded | 6a02b47adbc836b87d824c13 |
| w/ SFT\* | 83.33 | Not recorded | 6a02b47adbc836b87d824c13 |
| few-shot | 55.84 | Not recorded | 6a02b47adbc836b87d824c13 |
| few-shot | 55.61 | Not recorded | 6a02b47adbc836b87d824c13 |
| zero-shot | 52.83 | Not recorded | 6a02b47adbc836b87d824c13 |
| zero-shot | 48.51 | Not recorded | 6a02b47adbc836b87d824c13 |

### DPUA

Accuracy: 85.57. Result ID: 6a034a2a47f34ce924e880c8.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b47adbc836b87d824c13.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b47adbc836b87d824c13.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-3-Instruct (8B)"}. This record does not assert independent reproduction.

### DPUA

Accuracy: 85.57. Result ID: 6a034a2a47f34ce924e880cf.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b47adbc836b87d824c13.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b47adbc836b87d824c13.md>)

Recorded evaluation configuration: {"Backbone":"LLaMa-3.1-Instruct (8B)"}. This record does not assert independent reproduction.

### w/ SFT

Accuracy: 85.12. Result ID: 6a034a2a47f34ce924e880cb.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b47adbc836b87d824c13.md>)

Recorded evaluation configuration: {"Backbone":"LLaMa-3.1-Instruct (8B)"}. This record does not assert independent reproduction.

### w/ GRPO

Accuracy: 85.12. Result ID: 6a034a2a47f34ce924e880cd.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b47adbc836b87d824c13.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73101e209131cea5a395.md>)

[Implementation](<https://github.com/deepseek-ai/DeepSeek-Math>)

Recorded evaluation configuration: {"Backbone":"LLaMa-3.1-Instruct (8B)"}. This record does not assert independent reproduction.

### w/ GRPO\*

Accuracy: 84.97. Result ID: 6a034a2a47f34ce924e880ce.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b47adbc836b87d824c13.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73101e209131cea5a395.md>)

[Implementation](<https://github.com/deepseek-ai/DeepSeek-Math>)

Recorded evaluation configuration: {"Backbone":"LLaMa-3.1-Instruct (8B)"}. This record does not assert independent reproduction.

### w/ GRPO

Accuracy: 84.59. Result ID: 6a034a2a47f34ce924e880c6.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b47adbc836b87d824c13.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73101e209131cea5a395.md>)

[Implementation](<https://github.com/deepseek-ai/DeepSeek-Math>)

Recorded evaluation configuration: {"Backbone":"Qwen-3-Instruct (8B)"}. This record does not assert independent reproduction.

### w/ SFT

Accuracy: 84.38. Result ID: 6a034a2a47f34ce924e880c4.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b47adbc836b87d824c13.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-3-Instruct (8B)"}. This record does not assert independent reproduction.

### w/ GRPO\*

Accuracy: 84.08. Result ID: 6a034a2a47f34ce924e880c7.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b47adbc836b87d824c13.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73101e209131cea5a395.md>)

[Implementation](<https://github.com/deepseek-ai/DeepSeek-Math>)

Recorded evaluation configuration: {"Backbone":"Qwen-3-Instruct (8B)"}. This record does not assert independent reproduction.

### w/ SFT\*

Accuracy: 84.04. Result ID: 6a034a2a47f34ce924e880c5.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b47adbc836b87d824c13.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-3-Instruct (8B)"}. This record does not assert independent reproduction.

### w/ SFT\*

Accuracy: 83.33. Result ID: 6a034a2a47f34ce924e880cc.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b47adbc836b87d824c13.md>)

Recorded evaluation configuration: {"Backbone":"LLaMa-3.1-Instruct (8B)"}. This record does not assert independent reproduction.

### few-shot

Accuracy: 55.84. Result ID: 6a034a2a47f34ce924e880c3.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b47adbc836b87d824c13.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-3-Instruct (8B)"}. This record does not assert independent reproduction.

### few-shot

Accuracy: 55.61. Result ID: 6a034a2a47f34ce924e880ca.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b47adbc836b87d824c13.md>)

Recorded evaluation configuration: {"Backbone":"LLaMa-3.1-Instruct (8B)"}. This record does not assert independent reproduction.

### zero-shot

Accuracy: 52.83. Result ID: 6a034a2a47f34ce924e880c9.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b47adbc836b87d824c13.md>)

Recorded evaluation configuration: {"Backbone":"LLaMa-3.1-Instruct (8B)"}. This record does not assert independent reproduction.

### zero-shot

Accuracy: 48.51. Result ID: 6a034a2a47f34ce924e880c2.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b47adbc836b87d824c13.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-3-Instruct (8B)"}. This record does not assert independent reproduction.

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