# Readback Anomaly Detection on APCP

Source: [Readback Anomaly Detection on APCP](<https://www.sota2.com/research/sota/readback-anomaly-detection-on-apcp>)

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

Data updated: 2026-05-29T09:13:15.161Z

Evaluation of Readback Anomaly Detection performance on the APCP dataset, reporting metrics such as Accuracy and 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 | Readback Anomaly Detection |
| Dataset (including any stated split) | APCP |
| Metric | Accuracy |
| 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

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

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

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| SCOPE | 91.05 | Not recorded | 6a191d80dbd7150506f82759 |
| SCOPE | 90.61 | Not recorded | 6a191d80dbd7150506f82759 |
| SCOPE | 89.56 | Not recorded | 6a191d80dbd7150506f82759 |
| SCOPE | 88.51 | Not recorded | 6a191d80dbd7150506f82759 |
| DeBERTa | 87.02 | Not recorded | 6a191d80dbd7150506f82759 |
| SCOPE | 86.69 | Not recorded | 6a191d80dbd7150506f82759 |
| SuperICL | 84.12 | Not recorded | 6a191d80dbd7150506f82759 |
| SuperICL | 83.86 | Not recorded | 6a191d80dbd7150506f82759 |
| SuperICL | 80.88 | Not recorded | 6a191d80dbd7150506f82759 |
| SuperICL | 80.88 | Not recorded | 6a191d80dbd7150506f82759 |
| MilTOOD | 78.33 | Not recorded | 6a191d80dbd7150506f82759 |
| DICL | 70.96 | Not recorded | 6a191d80dbd7150506f82759 |
| DICL | 69.47 | Not recorded | 6a191d80dbd7150506f82759 |
| DICL | 67.98 | Not recorded | 6a191d80dbd7150506f82759 |
| ICL | 67.81 | Not recorded | 6a191d80dbd7150506f82759 |
| DICL | 67.19 | Not recorded | 6a191d80dbd7150506f82759 |
| GenICL | 66.93 | Not recorded | 6a191d80dbd7150506f82759 |
| GenICL | 66.58 | Not recorded | 6a191d80dbd7150506f82759 |
| SuperICL | 66.23 | Not recorded | 6a191d80dbd7150506f82759 |
| ICL | 66.05 | Not recorded | 6a191d80dbd7150506f82759 |
| GenICL | 65.88 | Not recorded | 6a191d80dbd7150506f82759 |
| ICL | 65.7 | Not recorded | 6a191d80dbd7150506f82759 |
| ICL | 65.61 | Not recorded | 6a191d80dbd7150506f82759 |
| GenICL | 64.74 | Not recorded | 6a191d80dbd7150506f82759 |
| ICL | 47.98 | Not recorded | 6a191d80dbd7150506f82759 |
| DICL | 47.98 | Not recorded | 6a191d80dbd7150506f82759 |
| GenICL | 47.98 | Not recorded | 6a191d80dbd7150506f82759 |
| Manual | 42.63 | Not recorded | 6a191d80dbd7150506f82759 |
| N-gram | 25.96 | Not recorded | 6a191d80dbd7150506f82759 |
| Regex | 25.61 | Not recorded | 6a191d80dbd7150506f82759 |

### SCOPE

Accuracy: 91.05. Result ID: 6a195701fbc373592c13e302.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"4-shot"}. This record does not assert independent reproduction.

### SCOPE

Accuracy: 90.61. Result ID: 6a195701fbc373592c13e301.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"3-shot"}. This record does not assert independent reproduction.

### SCOPE

Accuracy: 89.56. Result ID: 6a195701fbc373592c13e300.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"2-shot"}. This record does not assert independent reproduction.

### SCOPE

Accuracy: 88.51. Result ID: 6a195701fbc373592c13e2ff.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"1-shot"}. This record does not assert independent reproduction.

### DeBERTa

Accuracy: 87.02. Result ID: 6a195701fbc373592c13e2e7.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6aef1e209131cea2b2a6.md>)

[Implementation](<https://github.com/microsoft/DeBERTa111>)

Recorded evaluation configuration: {"Type":"Supervised Learning","Shots":"Overall"}. This record does not assert independent reproduction.

### SCOPE

Accuracy: 86.69. Result ID: 6a195701fbc373592c13e2fe.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"0-shot"}. This record does not assert independent reproduction.

### SuperICL

Accuracy: 84.12. Result ID: 6a195701fbc373592c13e2fd.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f581e209131cea4526e.md>)

[Implementation](<https://aka.ms/SuperICL>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"4-shot"}. This record does not assert independent reproduction.

### SuperICL

Accuracy: 83.86. Result ID: 6a195701fbc373592c13e2fc.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f581e209131cea4526e.md>)

[Implementation](<https://aka.ms/SuperICL>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"3-shot"}. This record does not assert independent reproduction.

### SuperICL

Accuracy: 80.88. Result ID: 6a195701fbc373592c13e2fa.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f581e209131cea4526e.md>)

[Implementation](<https://aka.ms/SuperICL>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"1-shot"}. This record does not assert independent reproduction.

### SuperICL

Accuracy: 80.88. Result ID: 6a195701fbc373592c13e2fb.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f581e209131cea4526e.md>)

[Implementation](<https://aka.ms/SuperICL>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"2-shot"}. This record does not assert independent reproduction.

### MilTOOD

Accuracy: 78.33. Result ID: 6a195701fbc373592c13e2e8.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72471e209131cea5580d.md>)

Recorded evaluation configuration: {"Type":"Supervised Learning","Shots":"Overall"}. This record does not assert independent reproduction.

### DICL

Accuracy: 70.96. Result ID: 6a195701fbc373592c13e2f2.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78691e209131cea784ea.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"3-shot"}. This record does not assert independent reproduction.

### DICL

Accuracy: 69.47. Result ID: 6a195701fbc373592c13e2f1.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78691e209131cea784ea.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"2-shot"}. This record does not assert independent reproduction.

### DICL

Accuracy: 67.98. Result ID: 6a195701fbc373592c13e2f0.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78691e209131cea784ea.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"1-shot"}. This record does not assert independent reproduction.

### ICL

Accuracy: 67.81. Result ID: 6a195701fbc373592c13e2ed.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa680c1e209131cea19d53.md>)

[Implementation](<https://github.com/openai/gpt-3/blob/master/overlap_frequency.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"3-shot"}. This record does not assert independent reproduction.

### DICL

Accuracy: 67.19. Result ID: 6a195701fbc373592c13e2f3.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78691e209131cea784ea.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"4-shot"}. This record does not assert independent reproduction.

### GenICL

Accuracy: 66.93. Result ID: 6a195701fbc373592c13e2f6.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78c21e209131cea7a598.md>)

[Implementation](<https://foundation-model-research.github.io/GenICL>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"2-shot"}. This record does not assert independent reproduction.

### GenICL

Accuracy: 66.58. Result ID: 6a195701fbc373592c13e2f7.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78c21e209131cea7a598.md>)

[Implementation](<https://foundation-model-research.github.io/GenICL>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"3-shot"}. This record does not assert independent reproduction.

### SuperICL

Accuracy: 66.23. Result ID: 6a195701fbc373592c13e2f9.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f581e209131cea4526e.md>)

[Implementation](<https://aka.ms/SuperICL>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"0-shot"}. This record does not assert independent reproduction.

### ICL

Accuracy: 66.05. Result ID: 6a195701fbc373592c13e2ec.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa680c1e209131cea19d53.md>)

[Implementation](<https://github.com/openai/gpt-3/blob/master/overlap_frequency.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"2-shot"}. This record does not assert independent reproduction.

### GenICL

Accuracy: 65.88. Result ID: 6a195701fbc373592c13e2f5.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78c21e209131cea7a598.md>)

[Implementation](<https://foundation-model-research.github.io/GenICL>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"1-shot"}. This record does not assert independent reproduction.

### ICL

Accuracy: 65.7. Result ID: 6a195701fbc373592c13e2ee.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa680c1e209131cea19d53.md>)

[Implementation](<https://github.com/openai/gpt-3/blob/master/overlap_frequency.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"4-shot"}. This record does not assert independent reproduction.

### ICL

Accuracy: 65.61. Result ID: 6a195701fbc373592c13e2eb.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa680c1e209131cea19d53.md>)

[Implementation](<https://github.com/openai/gpt-3/blob/master/overlap_frequency.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"1-shot"}. This record does not assert independent reproduction.

### GenICL

Accuracy: 64.74. Result ID: 6a195701fbc373592c13e2f8.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78c21e209131cea7a598.md>)

[Implementation](<https://foundation-model-research.github.io/GenICL>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"4-shot"}. This record does not assert independent reproduction.

### ICL

Accuracy: 47.98. Result ID: 6a195701fbc373592c13e2ea.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa680c1e209131cea19d53.md>)

[Implementation](<https://github.com/openai/gpt-3/blob/master/overlap_frequency.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"0-shot"}. This record does not assert independent reproduction.

### DICL

Accuracy: 47.98. Result ID: 6a195701fbc373592c13e2ef.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78691e209131cea784ea.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"0-shot"}. This record does not assert independent reproduction.

### GenICL

Accuracy: 47.98. Result ID: 6a195701fbc373592c13e2f4.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78c21e209131cea7a598.md>)

[Implementation](<https://foundation-model-research.github.io/GenICL>)

Recorded evaluation configuration: {"Backbone":"Qwen3-14B","Shots":"0-shot"}. This record does not assert independent reproduction.

### Manual

Accuracy: 42.63. Result ID: 6a195701fbc373592c13e2e9.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

Recorded evaluation configuration: {"Type":"LLM-based ICL","Backbone":"Qwen3-14B","Shots":"Overall"}. This record does not assert independent reproduction.

### N-gram

Accuracy: 25.96. Result ID: 6a195701fbc373592c13e2e6.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

Recorded evaluation configuration: {"Type":"Rule-based","Shots":"Overall"}. This record does not assert independent reproduction.

### Regex

Accuracy: 25.61. Result ID: 6a195701fbc373592c13e2e5.

[Reporting paper](<https://www.sota2.com/research/paper/6a191d80dbd7150506f82759.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e2a1e209131cea3e3c7.md>)

Recorded evaluation configuration: {"Type":"Rule-based","Shots":"Overall"}. This record does not assert independent reproduction.

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