# Short Text Clustering on AgNews

Source: [Short Text Clustering on AgNews](<https://www.sota2.com/research/sota/short-text-clustering-on-agnews>)

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

Data updated: 2026-02-26T05:21:25.132Z

Evaluation of Short Text Clustering performance on the AgNews dataset, measured using Accuracy (ACC) and Normalized Mutual Information (NMI).

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 | Short Text Clustering |
| Dataset (including any stated split) | AgNews |
| Metric | ACC |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 38 of 38 |
| 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

- [ACC](<https://www.sota2.com/research/sota/short-text-clustering-on-agnews.md?metric=ACC>): Higher is better; unit: PERCENT.

- [NMI](<https://www.sota2.com/research/sota/short-text-clustering-on-agnews.md?metric=NMI>): Higher is better; unit: SCALAR.

## Recorded results

| Method | ACC | Result date | Report paper |
| --- | --- | --- | --- |
| SCCL | 88.2 | Not recorded | 68fa6a0a1e209131cea25a86 |
| Gold | 85.4 | Not recorded | 68fa792f1e209131cea7c7c5 |
| RSTC-I | 85.39 | Not recorded | 68fa6f7f1e209131cea45fff |
| llm2vec+KMeans | 84.4 | Not recorded | 68fa792f1e209131cea7c7c5 |
| ZeroDL | 84.3 | Not recorded | 68fa792f1e209131cea7c7c5 |
| RSTC | 84.24 | Not recorded | 68fa6f7f1e209131cea45fff |
| SCCL | 83.1 | Not recorded | 68fa6f7f1e209131cea45fff |
| IDAS | 83.1 | Not recorded | 68fa792f1e209131cea7c7c5 |
| Gold | 82.9 | Not recorded | 68fa792f1e209131cea7c7c5 |
| Gold | 82.7 | Not recorded | 68fa792f1e209131cea7c7c5 |
| HAC-SD | 81.8 | Not recorded | 68fa6a0a1e209131cea25a86 |
| ZeroDL | 81.8 | Not recorded | 68fa792f1e209131cea7c7c5 |
| ZeroDL | 79.5 | Not recorded | 68fa792f1e209131cea7c7c5 |
| RSTC-C | 78.08 | Not recorded | 68fa6f7f1e209131cea45fff |
| ZeroDL | 75.2 | Not recorded | 68fa792f1e209131cea7c7c5 |
| ZeroDL | 74.5 | Not recorded | 68fa792f1e209131cea7c7c5 |
| ZeroDL | 70.9 | Not recorded | 68fa792f1e209131cea7c7c5 |
| Gold | 70.5 | Not recorded | 68fa792f1e209131cea7c7c5 |
| IDAS | 68.8 | Not recorded | 68fa792f1e209131cea7c7c5 |
| K-means\_IC | 66.3 | Not recorded | 68fa6f7f1e209131cea45fff |
| SBERT(k-means) | 65.95 | Not recorded | 68fa6f7f1e209131cea45fff |
| RSTC-OT | 65.94 | Not recorded | 68fa6f7f1e209131cea45fff |
| Gold | 62.2 | Not recorded | 68fa792f1e209131cea7c7c5 |
| Gold | 60.7 | Not recorded | 68fa792f1e209131cea7c7c5 |
| ZeroDL | 59.5 | Not recorded | 68fa792f1e209131cea7c7c5 |
| Gold | 59.4 | Not recorded | 68fa792f1e209131cea7c7c5 |
| IDAS | 44.6 | Not recorded | 68fa792f1e209131cea7c7c5 |
| IDAS | 38.1 | Not recorded | 68fa792f1e209131cea7c7c5 |
| IDAS | 35 | Not recorded | 68fa792f1e209131cea7c7c5 |
| TF-IDF | 34.5 | Not recorded | 68fa6a0a1e209131cea25a86 |
| TF-IDF | 34.39 | Not recorded | 68fa6f7f1e209131cea45fff |
| ZeroDL | 33.1 | Not recorded | 68fa792f1e209131cea7c7c5 |
| IDAS | 31.7 | Not recorded | 68fa792f1e209131cea7c7c5 |
| Gold | 31.4 | Not recorded | 68fa792f1e209131cea7c7c5 |
| IDAS | 29.4 | Not recorded | 68fa792f1e209131cea7c7c5 |
| BOW | 28.71 | Not recorded | 68fa6f7f1e209131cea45fff |
| BoW | 27.6 | Not recorded | 68fa6a0a1e209131cea25a86 |
| IDAS | 27.3 | Not recorded | 68fa792f1e209131cea7c7c5 |

### SCCL

ACC: 88.2. Result ID: 698e0272ffc3901de09a1b3c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0a1e209131cea25a86.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a0a1e209131cea25a86.md>)

[Implementation](<https://github.com/amazon-research/sccl>)

Recorded evaluation configuration: {"Averaging":"Average over five random runs"}. This record does not assert independent reproduction.

### Gold

ACC: 85.4. Result ID: 6995cf833badfaffe6d4502b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

Recorded evaluation configuration: {"Model":"gemma-2-27b-it","Prompt Order":"T-C"}. This record does not assert independent reproduction.

### RSTC-I

ACC: 85.39. Result ID: 6990c6683a6c314849032e74.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f7f1e209131cea45fff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f7f1e209131cea45fff.md>)

[Implementation](<https://github.com/hmllmh/RSTC>)

Recorded evaluation configuration: {"contrastive learning":"instance-wise","clustering":"k-means"}. This record does not assert independent reproduction.

### llm2vec+KMeans

ACC: 84.4. Result ID: 6995cf833badfaffe6d45019.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74901e209131cea62837.md>)

Recorded evaluation configuration: {"Model":"mistral-7b-it"}. This record does not assert independent reproduction.

### ZeroDL

ACC: 84.3. Result ID: 6995cf833badfaffe6d45023.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

Recorded evaluation configuration: {"Model":"Qwen-2.5-7b-it","Prompt Order":"T-C"}. This record does not assert independent reproduction.

### RSTC

ACC: 84.24. Result ID: 6990c6683a6c314849032e75.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f7f1e209131cea45fff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f7f1e209131cea45fff.md>)

[Implementation](<https://github.com/hmllmh/RSTC>)

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

### SCCL

ACC: 83.1. Result ID: 6990c6683a6c314849032e70.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f7f1e209131cea45fff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63671e209131ce9fe711.md>)

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

### IDAS

ACC: 83.1. Result ID: 6995cf833badfaffe6d45024.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f991e209131cea46883.md>)

Recorded evaluation configuration: {"Model":"Qwen-2.5-7b-it","Prompt Order":"T-C"}. This record does not assert independent reproduction.

### Gold

ACC: 82.9. Result ID: 6995cf833badfaffe6d45025.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

Recorded evaluation configuration: {"Model":"Qwen-2.5-7b-it","Prompt Order":"T-C"}. This record does not assert independent reproduction.

### Gold

ACC: 82.7. Result ID: 6995cf833badfaffe6d4501f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

Recorded evaluation configuration: {"Model":"mistral-7b-it","Prompt Order":"T-C"}. This record does not assert independent reproduction.

### HAC-SD

ACC: 81.8. Result ID: 698e0272ffc3901de09a1b3b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0a1e209131cea25a86.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67021e209131cea13bbf.md>)

Recorded evaluation configuration: {"Algorithm":"Hierarchical Agglomerative Clustering"}. This record does not assert independent reproduction.

### ZeroDL

ACC: 81.8. Result ID: 6995cf833badfaffe6d45029.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

Recorded evaluation configuration: {"Model":"gemma-2-27b-it","Prompt Order":"T-C"}. This record does not assert independent reproduction.

### ZeroDL

ACC: 79.5. Result ID: 6995cf833badfaffe6d4501d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

Recorded evaluation configuration: {"Model":"mistral-7b-it","Prompt Order":"T-C"}. This record does not assert independent reproduction.

### RSTC-C

ACC: 78.08. Result ID: 6990c6683a6c314849032e73.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f7f1e209131cea45fff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f7f1e209131cea45fff.md>)

[Implementation](<https://github.com/hmllmh/RSTC>)

Recorded evaluation configuration: {"contrastive learning":"class-wise","pseudo-labels":"SAOT"}. This record does not assert independent reproduction.

### ZeroDL

ACC: 75.2. Result ID: 6995cf833badfaffe6d4501a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

Recorded evaluation configuration: {"Model":"mistral-7b-it","Prompt Order":"C-T"}. This record does not assert independent reproduction.

### ZeroDL

ACC: 74.5. Result ID: 6995cf833badfaffe6d45026.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

Recorded evaluation configuration: {"Model":"gemma-2-27b-it","Prompt Order":"C-T"}. This record does not assert independent reproduction.

### ZeroDL

ACC: 70.9. Result ID: 6995cf833badfaffe6d45020.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

Recorded evaluation configuration: {"Model":"Qwen-2.5-7b-it","Prompt Order":"C-T"}. This record does not assert independent reproduction.

### Gold

ACC: 70.5. Result ID: 6995cf833badfaffe6d45028.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

Recorded evaluation configuration: {"Model":"gemma-2-27b-it","Prompt Order":"C-T"}. This record does not assert independent reproduction.

### IDAS

ACC: 68.8. Result ID: 6995cf833badfaffe6d45021.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f991e209131cea46883.md>)

Recorded evaluation configuration: {"Model":"Qwen-2.5-7b-it","Prompt Order":"C-T"}. This record does not assert independent reproduction.

### K-means\_IC

ACC: 66.3. Result ID: 6990c6683a6c314849032e6f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f7f1e209131cea45fff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67021e209131cea13bbf.md>)

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

### SBERT(k-means)

ACC: 65.95. Result ID: 6990c6683a6c314849032e71.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f7f1e209131cea45fff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66441e209131cea0f5ab.md>)

[Implementation](<https://github.com/UKPLab/sentence-transformers>)

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

### RSTC-OT

ACC: 65.94. Result ID: 6990c6683a6c314849032e72.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f7f1e209131cea45fff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f7f1e209131cea45fff.md>)

[Implementation](<https://github.com/hmllmh/RSTC>)

Recorded evaluation configuration: {"pseudo-labels":"traditional OT"}. This record does not assert independent reproduction.

### Gold

ACC: 62.2. Result ID: 6995cf833badfaffe6d45022.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

Recorded evaluation configuration: {"Model":"Qwen-2.5-7b-it","Prompt Order":"C-T"}. This record does not assert independent reproduction.

### Gold

ACC: 60.7. Result ID: 6995cf833badfaffe6d4501c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

Recorded evaluation configuration: {"Model":"mistral-7b-it","Prompt Order":"C-T"}. This record does not assert independent reproduction.

### ZeroDL

ACC: 59.5. Result ID: 6995cf833badfaffe6d45016.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

Recorded evaluation configuration: {"Model":"hcx-seed-0.5b-it","Prompt Order":"T-C"}. This record does not assert independent reproduction.

### Gold

ACC: 59.4. Result ID: 6995cf833badfaffe6d45018.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

Recorded evaluation configuration: {"Model":"hcx-seed-0.5b-it","Prompt Order":"T-C"}. This record does not assert independent reproduction.

### IDAS

ACC: 44.6. Result ID: 6995cf833badfaffe6d4501e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f991e209131cea46883.md>)

Recorded evaluation configuration: {"Model":"mistral-7b-it","Prompt Order":"T-C"}. This record does not assert independent reproduction.

### IDAS

ACC: 38.1. Result ID: 6995cf833badfaffe6d4501b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f991e209131cea46883.md>)

Recorded evaluation configuration: {"Model":"mistral-7b-it","Prompt Order":"C-T"}. This record does not assert independent reproduction.

### IDAS

ACC: 35. Result ID: 6995cf833badfaffe6d4502a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f991e209131cea46883.md>)

Recorded evaluation configuration: {"Model":"gemma-2-27b-it","Prompt Order":"T-C","Token Limit":"8192"}. This record does not assert independent reproduction.

### TF-IDF

ACC: 34.5. Result ID: 698e0272ffc3901de09a1b3a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0a1e209131cea25a86.md>)

Recorded evaluation configuration: {"Feature dimension":1500,"Algorithm":"K-means"}. This record does not assert independent reproduction.

### TF-IDF

ACC: 34.39. Result ID: 6990c6683a6c314849032e6e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f7f1e209131cea45fff.md>)

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

### ZeroDL

ACC: 33.1. Result ID: 6995cf833badfaffe6d45013.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

Recorded evaluation configuration: {"Model":"hcx-seed-0.5b-it","Prompt Order":"C-T"}. This record does not assert independent reproduction.

### IDAS

ACC: 31.7. Result ID: 6995cf833badfaffe6d45027.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f991e209131cea46883.md>)

Recorded evaluation configuration: {"Model":"gemma-2-27b-it","Prompt Order":"C-T","Token Limit":"8192"}. This record does not assert independent reproduction.

### Gold

ACC: 31.4. Result ID: 6995cf833badfaffe6d45015.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

Recorded evaluation configuration: {"Model":"hcx-seed-0.5b-it","Prompt Order":"C-T"}. This record does not assert independent reproduction.

### IDAS

ACC: 29.4. Result ID: 6995cf833badfaffe6d45014.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f991e209131cea46883.md>)

Recorded evaluation configuration: {"Model":"hcx-seed-0.5b-it","Prompt Order":"C-T","Token Limit":"8192"}. This record does not assert independent reproduction.

### BOW

ACC: 28.71. Result ID: 6990c6683a6c314849032e6d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f7f1e209131cea45fff.md>)

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

### BoW

ACC: 27.6. Result ID: 698e0272ffc3901de09a1b39.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0a1e209131cea25a86.md>)

Recorded evaluation configuration: {"Feature dimension":1500,"Algorithm":"K-means"}. This record does not assert independent reproduction.

### IDAS

ACC: 27.3. Result ID: 6995cf833badfaffe6d45017.

[Reporting paper](<https://www.sota2.com/research/paper/68fa792f1e209131cea7c7c5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f991e209131cea46883.md>)

Recorded evaluation configuration: {"Model":"hcx-seed-0.5b-it","Prompt Order":"T-C","Token Limit":"8192"}. This record does not assert independent reproduction.

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