# Language Model Accuracy Evaluation on LLM Evaluation Scenarios

Source: [Language Model Accuracy Evaluation on LLM Evaluation Scenarios](<https://www.sota2.com/research/sota/language-model-accuracy-evaluation-on-llm-evaluation-scenarios>)

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

Data updated: 2026-04-13T07:19:34.049Z

Evaluation of language model accuracy using the LLM Evaluation Scenarios dataset, focusing specifically on the reported Accuracy metric.

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 | Language Model Accuracy Evaluation |
| Dataset (including any stated split) | LLM Evaluation Scenarios |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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/language-model-accuracy-evaluation-on-llm-evaluation-scenarios.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| ICL Accuracy | 85.44 | Not recorded | 68fa7bb61e209131cea8a133 |
| ICL Accuracy | 81.93 | Not recorded | 68fa7bb61e209131cea8a133 |
| Accuracy with LTV | 81.37 | Not recorded | 68fa7bb61e209131cea8a133 |
| ICL Accuracy | 81.33 | Not recorded | 68fa7bb61e209131cea8a133 |
| Accuracy with LTV | 78.18 | Not recorded | 68fa7bb61e209131cea8a133 |
| Accuracy with LTV | 75.59 | Not recorded | 68fa7bb61e209131cea8a133 |
| Zero-Shot Accuracy | 14.82 | Not recorded | 68fa7bb61e209131cea8a133 |
| Zero-Shot Accuracy | 12.52 | Not recorded | 68fa7bb61e209131cea8a133 |
| Zero-Shot Accuracy | 2.51 | Not recorded | 68fa7bb61e209131cea8a133 |

### ICL Accuracy

Accuracy: 85.44. Result ID: 69dc985047f34ce924c32abc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb61e209131cea8a133.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: {"Model":"Qwen2.5-32B"}. This record does not assert independent reproduction.

### ICL Accuracy

Accuracy: 81.93. Result ID: 69dc985047f34ce924c32ab9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb61e209131cea8a133.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: {"Model":"Llama3-70B"}. This record does not assert independent reproduction.

### Accuracy with LTV

Accuracy: 81.37. Result ID: 69dc985047f34ce924c32ac0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb61e209131cea8a133.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bb61e209131cea8a133.md>)

Recorded evaluation configuration: {"Model":"Yi-34B","Injection Layer":"50% of model's total depth","Injection Position":"last-token hidden state"}. This record does not assert independent reproduction.

### ICL Accuracy

Accuracy: 81.33. Result ID: 69dc985047f34ce924c32abf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb61e209131cea8a133.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: {"Model":"Yi-34B"}. This record does not assert independent reproduction.

### Accuracy with LTV

Accuracy: 78.18. Result ID: 69dc985047f34ce924c32aba.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb61e209131cea8a133.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bb61e209131cea8a133.md>)

Recorded evaluation configuration: {"Model":"Llama3-70B","Injection Layer":"50% of model's total depth","Injection Position":"last-token hidden state"}. This record does not assert independent reproduction.

### Accuracy with LTV

Accuracy: 75.59. Result ID: 69dc985047f34ce924c32abd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb61e209131cea8a133.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bb61e209131cea8a133.md>)

Recorded evaluation configuration: {"Model":"Qwen2.5-32B","Injection Layer":"50% of model's total depth","Injection Position":"last-token hidden state"}. This record does not assert independent reproduction.

### Zero-Shot Accuracy

Accuracy: 14.82. Result ID: 69dc985047f34ce924c32abe.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb61e209131cea8a133.md>)

Recorded evaluation configuration: {"Model":"Yi-34B"}. This record does not assert independent reproduction.

### Zero-Shot Accuracy

Accuracy: 12.52. Result ID: 69dc985047f34ce924c32abb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb61e209131cea8a133.md>)

Recorded evaluation configuration: {"Model":"Qwen2.5-32B"}. This record does not assert independent reproduction.

### Zero-Shot Accuracy

Accuracy: 2.51. Result ID: 69dc985047f34ce924c32ab8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bb61e209131cea8a133.md>)

Recorded evaluation configuration: {"Model":"Llama3-70B"}. This record does not assert independent reproduction.

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