# Text-demo In-context Learning on Demo-ICL-Bench

Source: [Text-demo In-context Learning on Demo-ICL-Bench](<https://www.sota2.com/research/sota/text-demo-in-context-learning-on-demo-icl-bench>)

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

Data updated: 2026-02-26T05:34:14.475Z

Evaluation of Text-based In-Context Learning performance on the Demo-ICL-Bench dataset, measuring accuracy with and without demonstrations.

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-demo In-context Learning |
| Dataset (including any stated split) | Demo-ICL-Bench |
| Metric | Accuracy (Demo) |
| 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

- [Accuracy (Demo)](<https://www.sota2.com/research/sota/text-demo-in-context-learning-on-demo-icl-bench.md?metric=Accuracy%20(Demo)>): Higher is better; unit: PERCENT.

- [Accuracy (w/o Demo)](<https://www.sota2.com/research/sota/text-demo-in-context-learning-on-demo-icl-bench.md?metric=Accuracy%20(w%2Fo%20Demo)>): Higher is better; unit: PERCENT.

- [Delta ICL](<https://www.sota2.com/research/sota/text-demo-in-context-learning-on-demo-icl-bench.md?metric=Delta%20ICL>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (Demo) | Result date | Report paper |
| --- | --- | --- | --- |
| Human | 84 | Not recorded | 698abbe1c97a0b3e3b5de29d |
| Gemini-2.5-Pro | 54.4 | Not recorded | 698abbe1c97a0b3e3b5de29d |
| GPT-4o | 48.8 | Not recorded | 698abbe1c97a0b3e3b5de29d |
| Qwen2.5-VL | 45 | Not recorded | 698abbe1c97a0b3e3b5de29d |
| Demo-ICL | 43.4 | Not recorded | 698abbe1c97a0b3e3b5de29d |
| Demo-ICL (SFT) | 38.4 | Not recorded | 698abbe1c97a0b3e3b5de29d |
| VideoChat-R1 | 34.4 | Not recorded | 698abbe1c97a0b3e3b5de29d |
| Video-R1 | 33.6 | Not recorded | 698abbe1c97a0b3e3b5de29d |
| Qwen2.5-VL | 32.8 | Not recorded | 698abbe1c97a0b3e3b5de29d |
| Ola | 32.2 | Not recorded | 698abbe1c97a0b3e3b5de29d |
| InternVL-3 | 31.4 | Not recorded | 698abbe1c97a0b3e3b5de29d |
| Ola-Video (Base) | 31.4 | Not recorded | 698abbe1c97a0b3e3b5de29d |
| LLaVA-Video | 31 | Not recorded | 698abbe1c97a0b3e3b5de29d |
| Qwen2-VL | 29 | Not recorded | 698abbe1c97a0b3e3b5de29d |

### Human

Accuracy (Demo): 84. Result ID: 699726d35698665ec475602d.

[Reporting paper](<https://www.sota2.com/research/paper/698abbe1c97a0b3e3b5de29d.md>)

Recorded evaluation configuration: {"Size":null,"Frame":null}. This record does not assert independent reproduction.

### Gemini-2.5-Pro

Accuracy (Demo): 54.4. Result ID: 699726d35698665ec475602e.

[Reporting paper](<https://www.sota2.com/research/paper/698abbe1c97a0b3e3b5de29d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c311e209131cea8cef5.md>)

Recorded evaluation configuration: {"Size":null,"Frame":null}. This record does not assert independent reproduction.

### GPT-4o

Accuracy (Demo): 48.8. Result ID: 699726d35698665ec475602f.

[Reporting paper](<https://www.sota2.com/research/paper/698abbe1c97a0b3e3b5de29d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75681e209131cea67a98.md>)

Recorded evaluation configuration: {"Size":null,"Frame":null}. This record does not assert independent reproduction.

### Qwen2.5-VL

Accuracy (Demo): 45. Result ID: 699726d35698665ec4756037.

[Reporting paper](<https://www.sota2.com/research/paper/698abbe1c97a0b3e3b5de29d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa773a1e209131cea71d3c.md>)

Recorded evaluation configuration: {"Size":"72B","Frame":"32"}. This record does not assert independent reproduction.

### Demo-ICL

Accuracy (Demo): 43.4. Result ID: 699726d35698665ec475603a.

[Reporting paper](<https://www.sota2.com/research/paper/698abbe1c97a0b3e3b5de29d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/698abbe1c97a0b3e3b5de29d.md>)

[Implementation](<https://github.com/dongyh20/Demo-ICL>)

Recorded evaluation configuration: {"Size":"7B","Frame":"32"}. This record does not assert independent reproduction.

### Demo-ICL (SFT)

Accuracy (Demo): 38.4. Result ID: 699726d35698665ec4756039.

[Reporting paper](<https://www.sota2.com/research/paper/698abbe1c97a0b3e3b5de29d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/698abbe1c97a0b3e3b5de29d.md>)

[Implementation](<https://github.com/dongyh20/Demo-ICL>)

Recorded evaluation configuration: {"Size":"7B","Frame":"32"}. This record does not assert independent reproduction.

### VideoChat-R1

Accuracy (Demo): 34.4. Result ID: 699726d35698665ec4756036.

[Reporting paper](<https://www.sota2.com/research/paper/698abbe1c97a0b3e3b5de29d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7ba11e209131cea89767.md>)

Recorded evaluation configuration: {"Size":"7B","Frame":"32"}. This record does not assert independent reproduction.

### Video-R1

Accuracy (Demo): 33.6. Result ID: 699726d35698665ec4756035.

[Reporting paper](<https://www.sota2.com/research/paper/698abbe1c97a0b3e3b5de29d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c6f1e209131cea8e0ec.md>)

Recorded evaluation configuration: {"Size":"7B","Frame":"32"}. This record does not assert independent reproduction.

### Qwen2.5-VL

Accuracy (Demo): 32.8. Result ID: 699726d35698665ec4756033.

[Reporting paper](<https://www.sota2.com/research/paper/698abbe1c97a0b3e3b5de29d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa773a1e209131cea71d3c.md>)

Recorded evaluation configuration: {"Size":"7B","Frame":"32"}. This record does not assert independent reproduction.

### Ola

Accuracy (Demo): 32.2. Result ID: 699726d35698665ec4756031.

[Reporting paper](<https://www.sota2.com/research/paper/698abbe1c97a0b3e3b5de29d.md>)

Recorded evaluation configuration: {"Size":"7B","Frame":"32"}. This record does not assert independent reproduction.

### InternVL-3

Accuracy (Demo): 31.4. Result ID: 699726d35698665ec4756034.

[Reporting paper](<https://www.sota2.com/research/paper/698abbe1c97a0b3e3b5de29d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78241e209131cea76fa5.md>)

Recorded evaluation configuration: {"Size":"8B","Frame":"32"}. This record does not assert independent reproduction.

### Ola-Video (Base)

Accuracy (Demo): 31.4. Result ID: 699726d35698665ec4756038.

[Reporting paper](<https://www.sota2.com/research/paper/698abbe1c97a0b3e3b5de29d.md>)

Recorded evaluation configuration: {"Size":"7B","Frame":"32"}. This record does not assert independent reproduction.

### LLaVA-Video

Accuracy (Demo): 31. Result ID: 699726d35698665ec4756032.

[Reporting paper](<https://www.sota2.com/research/paper/698abbe1c97a0b3e3b5de29d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a561e209131cea82934.md>)

[Implementation](<https://llava-vl.github.io/blog/llava-video>)

Recorded evaluation configuration: {"Size":"7B","Frame":"32"}. This record does not assert independent reproduction.

### Qwen2-VL

Accuracy (Demo): 29. Result ID: 699726d35698665ec4756030.

[Reporting paper](<https://www.sota2.com/research/paper/698abbe1c97a0b3e3b5de29d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75151e209131cea65976.md>)

[Implementation](<https://github.com/QwenLM/Qwen2-VL>)

Recorded evaluation configuration: {"Size":"7B","Frame":"32"}. This record does not assert independent reproduction.

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