# Natural Language Inference on ChaosMNLI (test)

Source: [Natural Language Inference on ChaosMNLI (test)](<https://www.sota2.com/research/sota/natural-language-inference-on-chaosmnli-test>)

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

Data updated: 2026-02-26T05:47:37.230Z

Evaluation of Natural Language Inference capabilities using the test split of the ChaosMNLI dataset, measured by JSD and Accuracy.

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 | Natural Language Inference |
| Dataset (including any stated split) | ChaosMNLI (test) |
| Metric | JSD |
| Unit | SCALAR |
| Direction | Lower 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

- [JSD](<https://www.sota2.com/research/sota/natural-language-inference-on-chaosmnli-test.md?metric=JSD>): Lower is better; unit: SCALAR.

- [Accuracy](<https://www.sota2.com/research/sota/natural-language-inference-on-chaosmnli-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | JSD | Result date | Report paper |
| --- | --- | --- | --- |
| AmbiU | 0.2231 | Not recorded | 68fa6a221e209131cea263e0 |
| AmbiSMU | 0.2575 | Not recorded | 68fa6a221e209131cea263e0 |
| AmbiSM | 0.2619 | Not recorded | 68fa6a221e209131cea263e0 |
| AmbiSMU | 0.2748 | Not recorded | 68fa6a221e209131cea263e0 |
| AmbiU | 0.2843 | Not recorded | 68fa6a221e209131cea263e0 |
| AmbiSM | 0.3017 | Not recorded | 68fa6a221e209131cea263e0 |
| AmbiU | 0.3183 | Not recorded | 68fa6a221e209131cea263e0 |
| S/MNLI Baseline | 0.3349 | Not recorded | 68fa6a221e209131cea263e0 |
| AmbiSMU | 0.354 | Not recorded | 68fa6a221e209131cea263e0 |

### AmbiU

JSD: 0.2231. Result ID: 698e145bffc3901de09aa4ec.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a221e209131cea263e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a221e209131cea263e0.md>)

[Implementation](<https://github.com/mariomeissner/AmbiNLI>)

Recorded evaluation configuration: {"labels":"ambiguity distribution","filtering":"filtered"}. This record does not assert independent reproduction.

### AmbiSMU

JSD: 0.2575. Result ID: 698e145bffc3901de09aa4ee.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a221e209131cea263e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a221e209131cea263e0.md>)

[Implementation](<https://github.com/mariomeissner/AmbiNLI>)

Recorded evaluation configuration: {"labels":"ambiguity distribution"}. This record does not assert independent reproduction.

### AmbiSM

JSD: 0.2619. Result ID: 698e145bffc3901de09aa4e9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a221e209131cea263e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a221e209131cea263e0.md>)

[Implementation](<https://github.com/mariomeissner/AmbiNLI>)

Recorded evaluation configuration: {"labels":"ambiguity distribution"}. This record does not assert independent reproduction.

### AmbiSMU

JSD: 0.2748. Result ID: 698e145bffc3901de09aa4ef.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a221e209131cea263e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a221e209131cea263e0.md>)

[Implementation](<https://github.com/mariomeissner/AmbiNLI>)

Recorded evaluation configuration: {"labels":"ambiguity distribution","filtering":"filtered"}. This record does not assert independent reproduction.

### AmbiU

JSD: 0.2843. Result ID: 698e145bffc3901de09aa4eb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a221e209131cea263e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a221e209131cea263e0.md>)

[Implementation](<https://github.com/mariomeissner/AmbiNLI>)

Recorded evaluation configuration: {"labels":"ambiguity distribution"}. This record does not assert independent reproduction.

### AmbiSM

JSD: 0.3017. Result ID: 698e145bffc3901de09aa4e8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a221e209131cea263e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a221e209131cea263e0.md>)

[Implementation](<https://github.com/mariomeissner/AmbiNLI>)

Recorded evaluation configuration: {"labels":"gold"}. This record does not assert independent reproduction.

### AmbiU

JSD: 0.3183. Result ID: 698e145bffc3901de09aa4ea.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a221e209131cea263e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a221e209131cea263e0.md>)

[Implementation](<https://github.com/mariomeissner/AmbiNLI>)

Recorded evaluation configuration: {"labels":"gold"}. This record does not assert independent reproduction.

### S/MNLI Baseline

JSD: 0.3349. Result ID: 698e145bffc3901de09aa4e7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a221e209131cea263e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68821e209131cea1c8f8.md>)

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

### AmbiSMU

JSD: 0.354. Result ID: 698e145bffc3901de09aa4ed.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a221e209131cea263e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a221e209131cea263e0.md>)

[Implementation](<https://github.com/mariomeissner/AmbiNLI>)

Recorded evaluation configuration: {"labels":"gold"}. This record does not assert independent reproduction.

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