# Question Answering on AgriRegion-Eval 160 questions (test) Source: [Question Answering on AgriRegion-Eval 160 questions (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:41:37.277Z Evaluation of Question Answering performance on the test split of the AgriRegion-Eval 160 questions dataset, utilizing metrics such as EM, F1, BLEU, ROUGE, BERTScore, and RAGA. 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 | Question Answering | | Dataset (including any stated split) | AgriRegion-Eval 160 questions (test) | | Metric | EM | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 5 of 5 | | 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 - [EM](): Higher is better; unit: PERCENT. - [F1](): Higher is better; unit: PERCENT. - [BLEU-4](): Higher is better; unit: PERCENT. - [ROUGE-L](): Higher is better; unit: PERCENT. - [BERTScore](): Higher is better; unit: PERCENT. - [RAGA Precision](): Higher is better; unit: PERCENT. ## Recorded results | Method | EM | Result date | Report paper | | --- | --- | --- | --- | | AgriRegion | 76 | Not recorded | 693ba1936cfd488761a552de | | GPT-4-Turbo | 64 | Not recorded | 693ba1936cfd488761a552de | | Claude 3.5 Sonnet | 60 | Not recorded | 693ba1936cfd488761a552de | | Gemini 1.5 Pro | 56 | Not recorded | 693ba1936cfd488761a552de | | Mistral-7B-Instruct | 49 | Not recorded | 693ba1936cfd488761a552de | ### AgriRegion EM: 76. Result ID: 699629ef5698665ec46b16a5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Retrieval":"RAG","Top-k":"5"}. This record does not assert independent reproduction. ### GPT-4-Turbo EM: 64. Result ID: 699629ef5698665ec46b16a6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Claude 3.5 Sonnet EM: 60. Result ID: 699629ef5698665ec46b16a7. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Gemini 1.5 Pro EM: 56. Result ID: 699629ef5698665ec46b16a8. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Mistral-7B-Instruct EM: 49. Result ID: 699629ef5698665ec46b16a9. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.