# Vocabulary Difficulty Prediction on KVL data (test) Source: [Vocabulary Difficulty Prediction on KVL data (test)]() Collection index: [llms.txt]() Data updated: 2026-05-15T07:56:41.783Z Evaluation of vocabulary difficulty prediction performance on the KVL data test set, reporting Pearson Correlation Coefficients (PCC) across Chinese, German, and Spanish subsets. 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 data (test) | | Metric | PCC (Chinese) | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 17 of 17 | | 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 - [PCC (Chinese)](): Higher is better; unit: SCALAR. - [PCC (German)](): Higher is better; unit: SCALAR. - [PCC (Spanish)](): Higher is better; unit: SCALAR. - [PCC (Mean)](): Higher is better; unit: SCALAR. ## Recorded results | Method | PCC (Chinese) | Result date | Report paper | | --- | --- | --- | --- | | Statistical Optimum | 0.989 | Not recorded | 6a06a8a3dbc836b87d82a120 | | finetuned\_llms\_plus | 0.928 | Not recorded | 6a06a8a3dbc836b87d82a120 | | open\_max | 0.927 | Not recorded | 6a06a8a3dbc836b87d82a120 | | ≤32B LLM Average | 0.926 | Not recorded | 6a06a8a3dbc836b87d82a120 | | finetuned\_llms | 0.925 | Not recorded | 6a06a8a3dbc836b87d82a120 | | ≤14B LLM Average | 0.921 | Not recorded | 6a06a8a3dbc836b87d82a120 | | GLM-4-32B | 0.918 | Not recorded | 6a06a8a3dbc836b87d82a120 | | Qwen2.5-32B | 0.917 | Not recorded | 6a06a8a3dbc836b87d82a120 | | ≤9B LLM Average | 0.916 | Not recorded | 6a06a8a3dbc836b87d82a120 | | Ministral-3-14B | 0.915 | Not recorded | 6a06a8a3dbc836b87d82a120 | | closed\_max | 0.874 | Not recorded | 6a06a8a3dbc836b87d82a120 | | explainable | 0.837 | Not recorded | 6a06a8a3dbc836b87d82a120 | | explainable | 0.82 | Not recorded | 6a06a8a3dbc836b87d82a120 | | explainable | 0.815 | Not recorded | 6a06a8a3dbc836b87d82a120 | | Open-Track Baseline | 0.804 | Not recorded | 6a06a8a3dbc836b87d82a120 | | traditional | 0.767 | Not recorded | 6a06a8a3dbc836b87d82a120 | | Closed-Track Baseline | 0.753 | Not recorded | 6a06a8a3dbc836b87d82a120 | ### Statistical Optimum PCC (Chinese): 0.989. Result ID: 6a06d11b47f34ce924ee392f. [Reporting paper]() Recorded evaluation configuration: {"Category":"Upper Bound / Simulation"}. This record does not assert independent reproduction. ### finetuned\_llms\_plus PCC (Chinese): 0.928. Result ID: 6a06d11b47f34ce924ee3926. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Track":"Open-track submission","Fine-tuning":"Soft-target cross-entropy"}. This record does not assert independent reproduction. ### open\_max PCC (Chinese): 0.927. Result ID: 6a06d11b47f34ce924ee3925. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Track":"Open-track submission","Fine-tuning":"Soft-target cross-entropy"}. This record does not assert independent reproduction. ### ≤32B LLM Average PCC (Chinese): 0.926. Result ID: 6a06d11b47f34ce924ee3928. [Reporting paper]() Recorded evaluation configuration: {"Model Type":"Ensemble"}. This record does not assert independent reproduction. ### finetuned\_llms PCC (Chinese): 0.925. Result ID: 6a06d11b47f34ce924ee3927. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Track":"Open-track submission","Fine-tuning":"Soft-target cross-entropy"}. This record does not assert independent reproduction. ### ≤14B LLM Average PCC (Chinese): 0.921. Result ID: 6a06d11b47f34ce924ee392c. [Reporting paper]() Recorded evaluation configuration: {"Model Type":"Ensemble"}. This record does not assert independent reproduction. ### GLM-4-32B PCC (Chinese): 0.918. Result ID: 6a06d11b47f34ce924ee3929. [Reporting paper]() Recorded evaluation configuration: {"Model Size":"32B"}. This record does not assert independent reproduction. ### Qwen2.5-32B PCC (Chinese): 0.917. Result ID: 6a06d11b47f34ce924ee392a. [Reporting paper]() Recorded evaluation configuration: {"Model Size":"32B"}. This record does not assert independent reproduction. ### ≤9B LLM Average PCC (Chinese): 0.916. Result ID: 6a06d11b47f34ce924ee392d. [Reporting paper]() Recorded evaluation configuration: {"Model Type":"Ensemble"}. This record does not assert independent reproduction. ### Ministral-3-14B PCC (Chinese): 0.915. Result ID: 6a06d11b47f34ce924ee392b. [Reporting paper]() Recorded evaluation configuration: {"Model Size":"14B"}. This record does not assert independent reproduction. ### closed\_max PCC (Chinese): 0.874. Result ID: 6a06d11b47f34ce924ee3914. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"track":"closed-track"}. This record does not assert independent reproduction. ### explainable PCC (Chinese): 0.837. Result ID: 6a06d11b47f34ce924ee3915. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"track":"closed-track"}. This record does not assert independent reproduction. ### explainable PCC (Chinese): 0.82. Result ID: 6a06d11b47f34ce924ee3917. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"inference\_type":"std. inference"}. This record does not assert independent reproduction. ### explainable PCC (Chinese): 0.815. Result ID: 6a06d11b47f34ce924ee3918. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"inference\_type":"lin. regression"}. This record does not assert independent reproduction. ### Open-Track Baseline PCC (Chinese): 0.804. Result ID: 6a06d11b47f34ce924ee392e. [Reporting paper]() Recorded evaluation configuration: {"Category":"Shared Task Baseline"}. This record does not assert independent reproduction. ### traditional PCC (Chinese): 0.767. Result ID: 6a06d11b47f34ce924ee3916. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"track":"closed-track"}. This record does not assert independent reproduction. ### Closed-Track Baseline PCC (Chinese): 0.753. Result ID: 6a06d11b47f34ce924ee3919. [Reporting paper]() Recorded evaluation configuration: {"track":"closed-track"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.