# Classification (8 classes) on Twitter customer-support requests

Source: [Classification (8 classes) on Twitter customer-support requests](<https://www.sota2.com/research/sota/classification-8-classes-on-twitter-customer-support-requests>)

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

Data updated: 2026-04-13T09:57:00.523Z

Evaluation of 8-class classification performance on the Twitter customer-support requests dataset, measured using Accuracy.

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 | Classification (8 classes) |
| Dataset (including any stated split) | Twitter customer-support requests |
| Metric | Accuracy (8 classes) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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 (8 classes)](<https://www.sota2.com/research/sota/classification-8-classes-on-twitter-customer-support-requests.md?metric=Accuracy%20(8%20classes)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (8 classes) | Result date | Report paper |
| --- | --- | --- | --- |
| DSPy GEPA | 96.88 | Not recorded | 69dc78a17d0172140ec78beb |
| AIR | 95.31 | Not recorded | 69dc78a17d0172140ec78beb |
| Fine-tuning | 90.63 | Not recorded | 69dc78a17d0172140ec78beb |
| TextGrad | 46.87 | Not recorded | 69dc78a17d0172140ec78beb |
| DSPy MIPROv2 | 31.25 | Not recorded | 69dc78a17d0172140ec78beb |
| DSPy BootstrapFewShot (#16) | 18.75 | Not recorded | 69dc78a17d0172140ec78beb |
| KNN | 9.37 | Not recorded | 69dc78a17d0172140ec78beb |
| Initial prompt | 0 | Not recorded | 69dc78a17d0172140ec78beb |

### DSPy GEPA

Accuracy (8 classes): 96.88. Result ID: 69dcbdd547f34ce924c4202e.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78a17d0172140ec78beb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81ee3.md>)

Recorded evaluation configuration: {"Train Tokens (Base Model)":"1636241 / 10261","Train Tokens (Reflection Model)":"107506 / 94565","Inference Tokens (Base Model)":"184569 / 879"}. This record does not assert independent reproduction.

### AIR

Accuracy (8 classes): 95.31. Result ID: 69dcbdd547f34ce924c42030.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78a17d0172140ec78beb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69dc78a17d0172140ec78beb.md>)

Recorded evaluation configuration: {"Train Tokens (Embed Model)":"7282 / -","Train Tokens (Base Model)":"54704 / 1392","Train Tokens (Reflection Model)":"10301 / 10981","Inference Tokens (Base Model)":"55392 / 174"}. This record does not assert independent reproduction.

### Fine-tuning

Accuracy (8 classes): 90.63. Result ID: 69dcbdd547f34ce924c4202b.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78a17d0172140ec78beb.md>)

Recorded evaluation configuration: {"Train Tokens (Base Model)":"54153 / 1362","Inference Tokens (Base Model)":"6752 / 179"}. This record does not assert independent reproduction.

### TextGrad

Accuracy (8 classes): 46.87. Result ID: 69dcbdd547f34ce924c4202f.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78a17d0172140ec78beb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73981e209131cea5d7cd.md>)

[Implementation](<https://github.com/zou-group/textgrad>)

Recorded evaluation configuration: {"Train Tokens (Base Model)":"501625 / 394","Train Tokens (Reflection Model)":"876468 / 348127","Inference Tokens (Base Model)":"475808 / 215"}. This record does not assert independent reproduction.

### DSPy MIPROv2

Accuracy (8 classes): 31.25. Result ID: 69dcbdd547f34ce924c4202d.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78a17d0172140ec78beb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75121e209131cea65828.md>)

Recorded evaluation configuration: {"Train Tokens (Base Model)":"1676737 / 23626","Inference Tokens (Base Model)":"75257 / 886"}. This record does not assert independent reproduction.

### DSPy BootstrapFewShot (#16)

Accuracy (8 classes): 18.75. Result ID: 69dcbdd547f34ce924c4202c.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78a17d0172140ec78beb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d01e209131cea4d60e.md>)

[Implementation](<https://github.com/stanfordnlp/dspy>)

Recorded evaluation configuration: {"Train Tokens (Base Model)":"48874 / 562","Inference Tokens (Base Model)":"80032 / 934"}. This record does not assert independent reproduction.

### KNN

Accuracy (8 classes): 9.37. Result ID: 69dcbdd547f34ce924c4202a.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78a17d0172140ec78beb.md>)

Recorded evaluation configuration: {"Inference Tokens (Base Model)":"9008 / 183"}. This record does not assert independent reproduction.

### Initial prompt

Accuracy (8 classes): 0. Result ID: 69dcbdd547f34ce924c42029.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78a17d0172140ec78beb.md>)

Recorded evaluation configuration: {"Inference Tokens (Base Model)":"6752 / 180"}. This record does not assert independent reproduction.

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