# Topic Classification on AGNews (Macro-F1)

Source: [Topic Classification on AGNews (Macro-F1)](<https://www.sota2.com/research/sota/topic-classification-on-agnews-macro-f1>)

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

Data updated: 2026-03-05T07:20:37.308Z

Evaluation of Topic Classification performance on the AGNews dataset, reporting Macro-F1 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 | Topic Classification |
| Dataset (including any stated split) | AGNews |
| Metric | Macro-F1 |
| 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

- [Macro-F1](<https://www.sota2.com/research/sota/topic-classification-on-agnews-macro-f1.md?metric=Macro-F1>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Macro-F1 | Result date | Report paper |
| --- | --- | --- | --- |
| DC | 88.07 | Not recorded | 68fa78e51e209131cea7b0dd |
| SC | 87.42 | Not recorded | 68fa78e51e209131cea7b0dd |
| SC | 87.09 | Not recorded | 68fa78e51e209131cea7b0dd |
| DC | 86.86 | Not recorded | 68fa78e51e209131cea7b0dd |
| SC | 86.84 | Not recorded | 68fa78e51e209131cea7b0dd |
| SC | 86.01 | Not recorded | 68fa78e51e209131cea7b0dd |
| SC | 85.34 | Not recorded | 68fa78e51e209131cea7b0dd |
| SC | 85.32 | Not recorded | 68fa78e51e209131cea7b0dd |
| CC | 85.3 | Not recorded | 68fa78e51e209131cea7b0dd |
| CC | 82.27 | Not recorded | 68fa78e51e209131cea7b0dd |
| Base LLM | 75.64 | Not recorded | 68fa78e51e209131cea7b0dd |
| Base LLM | 74.55 | Not recorded | 68fa78e51e209131cea7b0dd |
| Base LLM | 71.17 | Not recorded | 68fa78e51e209131cea7b0dd |
| Base LLM | 70.59 | Not recorded | 68fa78e51e209131cea7b0dd |
| BC | 69.02 | Not recorded | 68fa78e51e209131cea7b0dd |
| DC | 65.73 | Not recorded | 68fa78e51e209131cea7b0dd |
| DC | 65.54 | Not recorded | 68fa78e51e209131cea7b0dd |
| BC | 65.36 | Not recorded | 68fa78e51e209131cea7b0dd |
| CC | 64.66 | Not recorded | 68fa78e51e209131cea7b0dd |
| BC | 63.85 | Not recorded | 68fa78e51e209131cea7b0dd |
| BC | 63.78 | Not recorded | 68fa78e51e209131cea7b0dd |
| CC | 63.73 | Not recorded | 68fa78e51e209131cea7b0dd |
| BC | 63.41 | Not recorded | 68fa78e51e209131cea7b0dd |
| CC | 63.38 | Not recorded | 68fa78e51e209131cea7b0dd |
| Base LLM | 63.1 | Not recorded | 68fa78e51e209131cea7b0dd |
| DC | 62.94 | Not recorded | 68fa78e51e209131cea7b0dd |
| CC | 62.29 | Not recorded | 68fa78e51e209131cea7b0dd |
| DC | 62.07 | Not recorded | 68fa78e51e209131cea7b0dd |
| Base LLM | 62.06 | Not recorded | 68fa78e51e209131cea7b0dd |
| BC | 61.58 | Not recorded | 68fa78e51e209131cea7b0dd |

### DC

Macro-F1: 88.07. Result ID: 69a92eb86dbb7075b952259f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70371e209131cea4998c.md>)

[Implementation](<https://github.com/fywalter/label-bias>)

Recorded evaluation configuration: {"Model":"Qwen","k-shot":"16"}. This record does not assert independent reproduction.

### SC

Macro-F1: 87.42. Result ID: 69a92eb86dbb7075b95225ab.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

Recorded evaluation configuration: {"Model":"Mistral","k-shot":"16"}. This record does not assert independent reproduction.

### SC

Macro-F1: 87.09. Result ID: 69a92eb86dbb7075b95225a6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

Recorded evaluation configuration: {"Model":"Llama","k-shot":"16"}. This record does not assert independent reproduction.

### DC

Macro-F1: 86.86. Result ID: 69a92eb86dbb7075b95224af.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70371e209131cea4998c.md>)

[Implementation](<https://github.com/fywalter/label-bias>)

Recorded evaluation configuration: {"Model":"Qwen","Shots":"8-shot"}. This record does not assert independent reproduction.

### SC

Macro-F1: 86.84. Result ID: 69a92eb86dbb7075b95224bb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

Recorded evaluation configuration: {"Model":"Mistral","Shots":"8-shot"}. This record does not assert independent reproduction.

### SC

Macro-F1: 86.01. Result ID: 69a92eb86dbb7075b95224b6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

Recorded evaluation configuration: {"Model":"Llama","Shots":"8-shot"}. This record does not assert independent reproduction.

### SC

Macro-F1: 85.34. Result ID: 69a92eb86dbb7075b95225a1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

Recorded evaluation configuration: {"Model":"Qwen","k-shot":"16"}. This record does not assert independent reproduction.

### SC

Macro-F1: 85.32. Result ID: 69a92eb86dbb7075b95224b1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

Recorded evaluation configuration: {"Model":"Qwen","Shots":"8-shot"}. This record does not assert independent reproduction.

### CC

Macro-F1: 85.3. Result ID: 69a92eb86dbb7075b952259e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a201e209131cea261f0.md>)

Recorded evaluation configuration: {"Model":"Qwen","k-shot":"16"}. This record does not assert independent reproduction.

### CC

Macro-F1: 82.27. Result ID: 69a92eb86dbb7075b95224ae.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a201e209131cea261f0.md>)

Recorded evaluation configuration: {"Model":"Qwen","Shots":"8-shot"}. This record does not assert independent reproduction.

### Base LLM

Macro-F1: 75.64. Result ID: 69a92eb86dbb7075b95225a2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

Recorded evaluation configuration: {"Model":"Llama","k-shot":"16"}. This record does not assert independent reproduction.

### Base LLM

Macro-F1: 74.55. Result ID: 69a92eb86dbb7075b95224b2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

Recorded evaluation configuration: {"Model":"Llama","Shots":"8-shot"}. This record does not assert independent reproduction.

### Base LLM

Macro-F1: 71.17. Result ID: 69a92eb86dbb7075b95225a7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

Recorded evaluation configuration: {"Model":"Mistral","k-shot":"16"}. This record does not assert independent reproduction.

### Base LLM

Macro-F1: 70.59. Result ID: 69a92eb86dbb7075b95224b7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

Recorded evaluation configuration: {"Model":"Mistral","Shots":"8-shot"}. This record does not assert independent reproduction.

### BC

Macro-F1: 69.02. Result ID: 69a92eb86dbb7075b95225a0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75f51e209131cea6aa55.md>)

Recorded evaluation configuration: {"Model":"Qwen","k-shot":"16"}. This record does not assert independent reproduction.

### DC

Macro-F1: 65.73. Result ID: 69a92eb86dbb7075b95224b4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70371e209131cea4998c.md>)

[Implementation](<https://github.com/fywalter/label-bias>)

Recorded evaluation configuration: {"Model":"Llama","Shots":"8-shot"}. This record does not assert independent reproduction.

### DC

Macro-F1: 65.54. Result ID: 69a92eb86dbb7075b95225a4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70371e209131cea4998c.md>)

[Implementation](<https://github.com/fywalter/label-bias>)

Recorded evaluation configuration: {"Model":"Llama","k-shot":"16"}. This record does not assert independent reproduction.

### BC

Macro-F1: 65.36. Result ID: 69a92eb86dbb7075b95224b0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75f51e209131cea6aa55.md>)

Recorded evaluation configuration: {"Model":"Qwen","Shots":"8-shot"}. This record does not assert independent reproduction.

### CC

Macro-F1: 64.66. Result ID: 69a92eb86dbb7075b95224b3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a201e209131cea261f0.md>)

Recorded evaluation configuration: {"Model":"Llama","Shots":"8-shot"}. This record does not assert independent reproduction.

### BC

Macro-F1: 63.85. Result ID: 69a92eb86dbb7075b95225a5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75f51e209131cea6aa55.md>)

Recorded evaluation configuration: {"Model":"Llama","k-shot":"16"}. This record does not assert independent reproduction.

### BC

Macro-F1: 63.78. Result ID: 69a92eb86dbb7075b95224b5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75f51e209131cea6aa55.md>)

Recorded evaluation configuration: {"Model":"Llama","Shots":"8-shot"}. This record does not assert independent reproduction.

### CC

Macro-F1: 63.73. Result ID: 69a92eb86dbb7075b95225a3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a201e209131cea261f0.md>)

Recorded evaluation configuration: {"Model":"Llama","k-shot":"16"}. This record does not assert independent reproduction.

### BC

Macro-F1: 63.41. Result ID: 69a92eb86dbb7075b95225aa.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75f51e209131cea6aa55.md>)

Recorded evaluation configuration: {"Model":"Mistral","k-shot":"16"}. This record does not assert independent reproduction.

### CC

Macro-F1: 63.38. Result ID: 69a92eb86dbb7075b95225a8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a201e209131cea261f0.md>)

Recorded evaluation configuration: {"Model":"Mistral","k-shot":"16"}. This record does not assert independent reproduction.

### Base LLM

Macro-F1: 63.1. Result ID: 69a92eb86dbb7075b952259d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

Recorded evaluation configuration: {"Model":"Qwen","k-shot":"16"}. This record does not assert independent reproduction.

### DC

Macro-F1: 62.94. Result ID: 69a92eb86dbb7075b95225a9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70371e209131cea4998c.md>)

[Implementation](<https://github.com/fywalter/label-bias>)

Recorded evaluation configuration: {"Model":"Mistral","k-shot":"16"}. This record does not assert independent reproduction.

### CC

Macro-F1: 62.29. Result ID: 69a92eb86dbb7075b95224b8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a201e209131cea261f0.md>)

Recorded evaluation configuration: {"Model":"Mistral","Shots":"8-shot"}. This record does not assert independent reproduction.

### DC

Macro-F1: 62.07. Result ID: 69a92eb86dbb7075b95224b9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70371e209131cea4998c.md>)

[Implementation](<https://github.com/fywalter/label-bias>)

Recorded evaluation configuration: {"Model":"Mistral","Shots":"8-shot"}. This record does not assert independent reproduction.

### Base LLM

Macro-F1: 62.06. Result ID: 69a92eb86dbb7075b95224ad.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

Recorded evaluation configuration: {"Model":"Qwen","Shots":"8-shot"}. This record does not assert independent reproduction.

### BC

Macro-F1: 61.58. Result ID: 69a92eb86dbb7075b95224ba.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e51e209131cea7b0dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75f51e209131cea6aa55.md>)

Recorded evaluation configuration: {"Model":"Mistral","Shots":"8-shot"}. This record does not assert independent reproduction.

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