# Text Classification on Restaurant, AI, ACL, AGNews Continual Learning Sequence (test)

Source: [Text Classification on Restaurant, AI, ACL, AGNews Continual Learning Sequence (test)](<https://www.sota2.com/research/sota/text-classification-on-restaurant-ai-acl-agnews-continual-learning-sequence-test>)

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

Data updated: 2026-02-26T06:02:09.287Z

Evaluation of Text Classification performance across a continual learning sequence involving the Restaurant, AI, ACL, and AGNews datasets, using the test splits.

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 | Text Classification |
| Dataset (including any stated split) | Restaurant, AI, ACL, AGNews Continual Learning Sequence (test) |
| Metric | Restaurant MF1 |
| 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

- [Restaurant MF1](<https://www.sota2.com/research/sota/text-classification-on-restaurant-ai-acl-agnews-continual-learning-sequence-test.md?metric=Restaurant%20MF1>): Higher is better; unit: PERCENT.

- [Restaurant Acc](<https://www.sota2.com/research/sota/text-classification-on-restaurant-ai-acl-agnews-continual-learning-sequence-test.md?metric=Restaurant%20Acc>): Higher is better; unit: PERCENT.

- [AI MF1](<https://www.sota2.com/research/sota/text-classification-on-restaurant-ai-acl-agnews-continual-learning-sequence-test.md?metric=AI%20MF1>): Higher is better; unit: PERCENT.

- [AI Acc](<https://www.sota2.com/research/sota/text-classification-on-restaurant-ai-acl-agnews-continual-learning-sequence-test.md?metric=AI%20Acc>): Higher is better; unit: PERCENT.

- [ACL MF1](<https://www.sota2.com/research/sota/text-classification-on-restaurant-ai-acl-agnews-continual-learning-sequence-test.md?metric=ACL%20MF1>): Higher is better; unit: PERCENT.

- [ACL Acc](<https://www.sota2.com/research/sota/text-classification-on-restaurant-ai-acl-agnews-continual-learning-sequence-test.md?metric=ACL%20Acc>): Higher is better; unit: PERCENT.

- [AGNews MF1](<https://www.sota2.com/research/sota/text-classification-on-restaurant-ai-acl-agnews-continual-learning-sequence-test.md?metric=AGNews%20MF1>): Higher is better; unit: PERCENT.

- [AGNews Acc](<https://www.sota2.com/research/sota/text-classification-on-restaurant-ai-acl-agnews-continual-learning-sequence-test.md?metric=AGNews%20Acc>): Higher is better; unit: PERCENT.

- [Average MF1](<https://www.sota2.com/research/sota/text-classification-on-restaurant-ai-acl-agnews-continual-learning-sequence-test.md?metric=Average%20MF1>): Higher is better; unit: PERCENT.

- [Average Acc](<https://www.sota2.com/research/sota/text-classification-on-restaurant-ai-acl-agnews-continual-learning-sequence-test.md?metric=Average%20Acc>): Higher is better; unit: PERCENT.

- [Forget R. MF1](<https://www.sota2.com/research/sota/text-classification-on-restaurant-ai-acl-agnews-continual-learning-sequence-test.md?metric=Forget%20R.%20MF1>): Higher is better; unit: PERCENT.

- [Forget R. Acc](<https://www.sota2.com/research/sota/text-classification-on-restaurant-ai-acl-agnews-continual-learning-sequence-test.md?metric=Forget%20R.%20Acc>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Restaurant MF1 | Result date | Report paper |
| --- | --- | --- | --- |
| CPT | 53.9 | Not recorded | 68fa6d9d1e209131cea3acf8 |
| RoBERTa-ONE | 53.63 | Not recorded | 68fa6d9d1e209131cea3acf8 |
| DEMIX | 53.14 | Not recorded | 68fa6d9d1e209131cea3acf8 |
| Adapter-ONE | 52.19 | Not recorded | 68fa6d9d1e209131cea3acf8 |
| BCL | 51.7 | Not recorded | 68fa6d9d1e209131cea3acf8 |
| RoBERTa | 50.61 | Not recorded | 68fa6d9d1e209131cea3acf8 |
| HAT | 50.45 | Not recorded | 68fa6d9d1e209131cea3acf8 |
| EWC | 48.32 | Not recorded | 68fa6d9d1e209131cea3acf8 |
| DER++ | 48.09 | Not recorded | 68fa6d9d1e209131cea3acf8 |
| Adapter-NCL | 47.42 | Not recorded | 68fa6d9d1e209131cea3acf8 |
| Adapter | 45.4 | Not recorded | 68fa6d9d1e209131cea3acf8 |
| RoBERTa-NCL | 42.59 | Not recorded | 68fa6d9d1e209131cea3acf8 |
| KD | 39.75 | Not recorded | 68fa6d9d1e209131cea3acf8 |
| Prompt-ONE | 28.93 | Not recorded | 68fa6d9d1e209131cea3acf8 |

### CPT

Restaurant MF1: 53.9. Result ID: 698fc3e23a6c314849fbb0c7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d9d1e209131cea3acf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d9d1e209131cea3acf8.md>)

[Implementation](<https://github.com/UIC-Liu-Lab/CPT>)

Recorded evaluation configuration: {"Category":"CL","Domain Training Order":"Restaurant -\> AI -\> ACL -\> AGNews"}. This record does not assert independent reproduction.

### RoBERTa-ONE

Restaurant MF1: 53.63. Result ID: 698fc3e23a6c314849fbb0bc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d9d1e209131cea3acf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa677a1e209131cea16a90.md>)

Recorded evaluation configuration: {"Category":"Non-CL","Domain Training Order":"Restaurant -\> AI -\> ACL -\> AGNews"}. This record does not assert independent reproduction.

### DEMIX

Restaurant MF1: 53.14. Result ID: 698fc3e23a6c314849fbb0bf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d9d1e209131cea3acf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a981e209131cea29372.md>)

Recorded evaluation configuration: {"Category":"Non-CL","Domain Training Order":"Restaurant -\> AI -\> ACL -\> AGNews"}. This record does not assert independent reproduction.

### Adapter-ONE

Restaurant MF1: 52.19. Result ID: 698fc3e23a6c314849fbb0bd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d9d1e209131cea3acf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68fa1e209131cea1f79e.md>)

Recorded evaluation configuration: {"Category":"Non-CL","Domain Training Order":"Restaurant -\> AI -\> ACL -\> AGNews"}. This record does not assert independent reproduction.

### BCL

Restaurant MF1: 51.7. Result ID: 698fc3e23a6c314849fbb0c3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d9d1e209131cea3acf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b7f1e209131cea2e981.md>)

[Implementation](<https://github.com/ZixuanKe/PyContinual>)

Recorded evaluation configuration: {"Category":"CL","Domain Training Order":"Restaurant -\> AI -\> ACL -\> AGNews"}. This record does not assert independent reproduction.

### RoBERTa

Restaurant MF1: 50.61. Result ID: 698fc3e23a6c314849fbb0ba.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d9d1e209131cea3acf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66131e209131cea0e18b.md>)

[Implementation](<https://github.com/pytorch/fairseq>)

Recorded evaluation configuration: {"Category":"Non-CL","Domain Training Order":"Restaurant -\> AI -\> ACL -\> AGNews"}. This record does not assert independent reproduction.

### HAT

Restaurant MF1: 50.45. Result ID: 698fc3e23a6c314849fbb0c2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d9d1e209131cea3acf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64681e209131cea0485c.md>)

[Implementation](<https://github.com/joansj/hat>)

Recorded evaluation configuration: {"Category":"CL","Domain Training Order":"Restaurant -\> AI -\> ACL -\> AGNews"}. This record does not assert independent reproduction.

### EWC

Restaurant MF1: 48.32. Result ID: 698fc3e23a6c314849fbb0c5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d9d1e209131cea3acf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68a11e209131cea1d3e2.md>)

[Implementation](<https://github.com/aimagelab/mammoth>)

Recorded evaluation configuration: {"Category":"CL","Domain Training Order":"Restaurant -\> AI -\> ACL -\> AGNews"}. This record does not assert independent reproduction.

### DER++

Restaurant MF1: 48.09. Result ID: 698fc3e23a6c314849fbb0c6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d9d1e209131cea3acf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68a11e209131cea1d3e2.md>)

[Implementation](<https://github.com/aimagelab/mammoth>)

Recorded evaluation configuration: {"Category":"CL","Domain Training Order":"Restaurant -\> AI -\> ACL -\> AGNews"}. This record does not assert independent reproduction.

### Adapter-NCL

Restaurant MF1: 47.42. Result ID: 698fc3e23a6c314849fbb0c1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d9d1e209131cea3acf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65e31e209131cea0d324.md>)

[Implementation](<https://github.com/google-research/adapter-bert>)

Recorded evaluation configuration: {"Category":"CL","Domain Training Order":"Restaurant -\> AI -\> ACL -\> AGNews"}. This record does not assert independent reproduction.

### Adapter

Restaurant MF1: 45.4. Result ID: 698fc3e23a6c314849fbb0bb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d9d1e209131cea3acf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65e31e209131cea0d324.md>)

[Implementation](<https://github.com/google-research/adapter-bert>)

Recorded evaluation configuration: {"Category":"Non-CL","Domain Training Order":"Restaurant -\> AI -\> ACL -\> AGNews"}. This record does not assert independent reproduction.

### RoBERTa-NCL

Restaurant MF1: 42.59. Result ID: 698fc3e23a6c314849fbb0c0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d9d1e209131cea3acf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa677a1e209131cea16a90.md>)

Recorded evaluation configuration: {"Category":"CL","Domain Training Order":"Restaurant -\> AI -\> ACL -\> AGNews"}. This record does not assert independent reproduction.

### KD

Restaurant MF1: 39.75. Result ID: 698fc3e23a6c314849fbb0c4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d9d1e209131cea3acf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa61b81e209131ce9f5b87.md>)

Recorded evaluation configuration: {"Category":"CL","Domain Training Order":"Restaurant -\> AI -\> ACL -\> AGNews"}. This record does not assert independent reproduction.

### Prompt-ONE

Restaurant MF1: 28.93. Result ID: 698fc3e23a6c314849fbb0be.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d9d1e209131cea3acf8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6aa51e209131cea296c9.md>)

Recorded evaluation configuration: {"Category":"Non-CL","Domain Training Order":"Restaurant -\> AI -\> ACL -\> AGNews"}. This record does not assert independent reproduction.

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