Spaces:
Sleeping
Sleeping
| enabled: true | |
| module_name: 基本教育 | |
| runtime_python: python | |
| runtime_cli_module: elbench.basic_education_runtime.cli.main | |
| command_timeout_seconds: 14400 | |
| continue_on_error: true | |
| workspace_root: outputs/basic_education | |
| test_endpoint_model_id: qwen-test | |
| evaluator_model_id: null | |
| default_pass_threshold: 3.0 | |
| scenarios: | |
| - scenario_id: knowledge_point_explanation | |
| source_file: knowledge.yaml | |
| subset: 知识点讲解 | |
| task: basic_knowledge_point_explanation | |
| template_path: 基本教育/知识点讲解/knowledge.yaml | |
| expected_tasks: 10 | |
| multi_turn: false | |
| dimension: 知识点讲解 | |
| target_model_keys: [4o_teacher] | |
| test_endpoint_model_keys: [] | |
| judge_model_keys: [evaler] | |
| pass_threshold: 3.0 | |
| metadata: | |
| topic: basic_education | |
| scene: knowledge | |
| - scenario_id: contextualized_question_generation | |
| source_file: question.yaml | |
| subset: 情景化出题 | |
| task: basic_contextualized_question_generation | |
| template_path: 基本教育/情景化出题/question.yaml | |
| expected_tasks: 10 | |
| multi_turn: false | |
| dimension: 情景化出题 | |
| target_model_keys: [test_model] | |
| test_endpoint_model_keys: [] | |
| judge_model_keys: [judge_model] | |
| pass_threshold: 3.0 | |
| metadata: | |
| topic: basic_education | |
| scene: question | |
| - scenario_id: interdisciplinary_lesson_plan_generation | |
| source_file: cross.yaml | |
| subset: 跨学科教案生成 | |
| task: basic_interdisciplinary_lesson_plan_generation | |
| template_path: 基本教育/跨学科教案生成/cross.yaml | |
| expected_tasks: 10 | |
| multi_turn: false | |
| dimension: 跨学科教案生成 | |
| target_model_keys: [test_model] | |
| test_endpoint_model_keys: [] | |
| judge_model_keys: [judge_model] | |
| pass_threshold: 3.0 | |
| metadata: | |
| topic: basic_education | |
| scene: cross | |
| - scenario_id: guided_problem_solving_teaching | |
| source_file: config_guided_task.yaml | |
| subset: 引导式讲题 | |
| task: basic_guided_problem_solving_teaching | |
| template_path: 基本教育/引导式讲题/config_guided_task.yaml | |
| expected_tasks: 15 | |
| multi_turn: true | |
| dimension: 引导式讲题 | |
| target_model_keys: [teacher] | |
| test_endpoint_model_keys: [stu] | |
| judge_model_keys: [eval] | |
| pass_threshold: 3.0 | |
| metadata: | |
| topic: basic_education | |
| scene: guided | |