# Vocabulary Difficulty Prediction on KVL BEA Shared Task Open Track 2026 (test) Source: [Vocabulary Difficulty Prediction on KVL BEA Shared Task Open Track 2026 (test)]() Collection index: [llms.txt]() Data updated: 2026-05-15T07:56:41.817Z Evaluation of Vocabulary Difficulty Prediction performance on the KVL BEA Shared Task Open Track test set, reporting language-specific RMSE scores for Chinese, German, and Spanish. 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 | Vocabulary Difficulty Prediction | | Dataset (including any stated split) | KVL BEA Shared Task Open Track 2026 (test) | | Metric | Chinese RMSE | | Unit | SCALAR | | Direction | Lower is better | | Results loaded | 11 of 11 | | 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 - [Chinese RMSE](): Lower is better; unit: SCALAR. - [German RMSE](): Lower is better; unit: SCALAR. - [Spanish RMSE](): Lower is better; unit: SCALAR. - [Mean RMSE](): Lower is better; unit: SCALAR. ## Recorded results | Method | Chinese RMSE | Result date | Report paper | | --- | --- | --- | --- | | Statistical Optimum | 0.321 | Not recorded | 6a06a8a3dbc836b87d82a120 | | finetuned\_llms\_plus | 0.63 | Not recorded | 6a06a8a3dbc836b87d82a120 | | open\_max | 0.631 | Not recorded | 6a06a8a3dbc836b87d82a120 | | finetuned\_llms | 0.64 | Not recorded | 6a06a8a3dbc836b87d82a120 | | ≤32B LLM Average | 0.645 | Not recorded | 6a06a8a3dbc836b87d82a120 | | ≤14B LLM Average | 0.662 | Not recorded | 6a06a8a3dbc836b87d82a120 | | GLM-4-32B | 0.678 | Not recorded | 6a06a8a3dbc836b87d82a120 | | Qwen2.5-32B | 0.678 | Not recorded | 6a06a8a3dbc836b87d82a120 | | Ministral-3-14B | 0.681 | Not recorded | 6a06a8a3dbc836b87d82a120 | | ≤9B LLM Average | 0.683 | Not recorded | 6a06a8a3dbc836b87d82a120 | | Open-Track Baseline | 1.034 | Not recorded | 6a06a8a3dbc836b87d82a120 | ### Statistical Optimum Chinese RMSE: 0.321. Result ID: 6a06d11b47f34ce924ee3924. [Reporting paper]() Recorded evaluation configuration: {"Type":"Simulation"}. This record does not assert independent reproduction. ### finetuned\_llms\_plus Chinese RMSE: 0.63. Result ID: 6a06d11b47f34ce924ee391b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Submission"}. This record does not assert independent reproduction. ### open\_max Chinese RMSE: 0.631. Result ID: 6a06d11b47f34ce924ee391a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Submission"}. This record does not assert independent reproduction. ### finetuned\_llms Chinese RMSE: 0.64. Result ID: 6a06d11b47f34ce924ee391c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Submission"}. This record does not assert independent reproduction. ### ≤32B LLM Average Chinese RMSE: 0.645. Result ID: 6a06d11b47f34ce924ee391d. [Reporting paper]() Recorded evaluation configuration: {"Type":"Average Ensemble","Size constraint":"≤32B"}. This record does not assert independent reproduction. ### ≤14B LLM Average Chinese RMSE: 0.662. Result ID: 6a06d11b47f34ce924ee3921. [Reporting paper]() Recorded evaluation configuration: {"Type":"Average Ensemble","Size constraint":"≤14B"}. This record does not assert independent reproduction. ### GLM-4-32B Chinese RMSE: 0.678. Result ID: 6a06d11b47f34ce924ee391e. [Reporting paper]() Recorded evaluation configuration: {"Model Family":"GLM-4","Size":"32B"}. This record does not assert independent reproduction. ### Qwen2.5-32B Chinese RMSE: 0.678. Result ID: 6a06d11b47f34ce924ee391f. [Reporting paper]() Recorded evaluation configuration: {"Model Family":"Qwen2.5","Size":"32B"}. This record does not assert independent reproduction. ### Ministral-3-14B Chinese RMSE: 0.681. Result ID: 6a06d11b47f34ce924ee3920. [Reporting paper]() Recorded evaluation configuration: {"Model Family":"Ministral-3","Size":"14B"}. This record does not assert independent reproduction. ### ≤9B LLM Average Chinese RMSE: 0.683. Result ID: 6a06d11b47f34ce924ee3922. [Reporting paper]() Recorded evaluation configuration: {"Type":"Average Ensemble","Size constraint":"≤9B"}. This record does not assert independent reproduction. ### Open-Track Baseline Chinese RMSE: 1.034. Result ID: 6a06d11b47f34ce924ee3923. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"fine-tuned XLM-RoBERTa"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.