# Natural Language Inference on AmericasNLI (test) Source: [Natural Language Inference on AmericasNLI (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:34:31.080Z Evaluation of Natural Language Inference performance on the test split of the AmericasNLI dataset, measured using a specific set of 12 language-specific metrics. 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) | AmericasNLI (test) | | Metric | Accuracy (aym) | | 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 (aym)](): Higher is better; unit: PERCENT. - [Accuracy (bzd)](): Higher is better; unit: PERCENT. - [Accuracy (cni)](): Higher is better; unit: PERCENT. - [Accuracy (gn)](): Higher is better; unit: PERCENT. - [Accuracy (hch)](): Higher is better; unit: PERCENT. - [Accuracy (nah)](): Higher is better; unit: PERCENT. - [Accuracy (oto)](): Higher is better; unit: PERCENT. - [Accuracy (quy)](): Higher is better; unit: PERCENT. - [Accuracy (shp)](): Higher is better; unit: PERCENT. - [Accuracy (tar)](): Higher is better; unit: PERCENT. - [Average Accuracy](): Higher is better; unit: PERCENT. - [Avg Change (Delta)](): Higher is better; unit: SCALAR. - [Accuracy (grn)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy (aym) | Result date | Report paper | | --- | --- | --- | --- | | BISTIL-TF | 58.9 | Not recorded | 68fa6f911e209131cea464e1 | | LTSFT | 58.1 | Not recorded | 68fa6f911e209131cea464e1 | | BISTIL-TF | 57.7 | Not recorded | 68fa6f911e209131cea464e1 | | UniBridge | 52.13 | Not recorded | 68fa748d1e209131cea626ca | | MAD-X | 39.46 | Not recorded | 68fa748d1e209131cea626ca | | XLM-R | 36.26 | Not recorded | 68fa748d1e209131cea626ca | | UniBridge | 35.73 | Not recorded | 68fa748d1e209131cea626ca | | mBERT | 33.33 | Not recorded | 68fa748d1e209131cea626ca | | MAD-X | 33.06 | Not recorded | 68fa748d1e209131cea626ca | ### BISTIL-TF Accuracy (aym): 58.9. Result ID: 6990e04d3a6c31484903f3fc. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"LRF":2,"mode":"Zero-shot"}. This record does not assert independent reproduction. ### LTSFT Accuracy (aym): 58.1. Result ID: 6990e04d3a6c31484903f3fb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"mode":"Zero-shot"}. This record does not assert independent reproduction. ### BISTIL-TF Accuracy (aym): 57.7. Result ID: 6990e04d3a6c31484903f3fd. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"LRF":3,"mode":"Zero-shot"}. This record does not assert independent reproduction. ### UniBridge Accuracy (aym): 52.13. Result ID: 69934b3e3a6c31484914fbd1. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"XLM-R"}. This record does not assert independent reproduction. ### MAD-X Accuracy (aym): 39.46. Result ID: 69934b3e3a6c31484914fbd0. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"XLM-R"}. This record does not assert independent reproduction. ### XLM-R Accuracy (aym): 36.26. Result ID: 69934b3e3a6c31484914fbcf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"XLM-R"}. This record does not assert independent reproduction. ### UniBridge Accuracy (aym): 35.73. Result ID: 69934b3e3a6c31484914fbce. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"mBERT"}. This record does not assert independent reproduction. ### mBERT Accuracy (aym): 33.33. Result ID: 69934b3e3a6c31484914fbcc. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"mBERT"}. This record does not assert independent reproduction. ### MAD-X Accuracy (aym): 33.06. Result ID: 69934b3e3a6c31484914fbcd. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"mBERT"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.