Upload coherence.jsonl

data/coherence.jsonl

@@ -0,0 +1,125 @@
+{"pair_id": "cohe_001", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_002", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_003", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_004", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_005", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_006", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_007", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_008", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_009", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_010", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_011", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_012", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_013", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_014", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_015", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_016", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_017", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_018", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_019", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_020", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_021", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_022", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_023", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_024", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_025", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "response_being_judged": "The meeting discussed budget allocation. Marketing needs increased funding. Sales performed well last quarter. The CEO approved the proposal.", "ground_truth_label": "score_1", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_026", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_027", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_028", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_029", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_030", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_031", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_032", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_033", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_034", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_035", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_036", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_037", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_038", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_039", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_040", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_041", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_042", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_043", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_044", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_045", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_046", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_047", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_048", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_049", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_050", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "response_being_judged": "Algorithm efficiency depends on complexity analysis. Big O notation measures worst-case performance. Different algorithms solve problems differently.", "ground_truth_label": "score_2", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_051", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_052", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_053", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_054", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_055", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_056", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_057", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_058", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_059", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_060", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_061", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_062", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_063", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_064", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_065", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_066", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_067", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_068", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_069", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_070", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_071", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_072", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_073", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_074", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_075", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "response_being_judged": "Climate change affects global temperatures. Ice caps are melting. We need renewable energy. Solar panels are expensive.", "ground_truth_label": "score_3", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_076", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_077", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_078", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_079", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_080", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_081", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_082", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_083", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_084", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_085", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_086", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_087", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_088", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_089", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_090", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_091", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_092", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_093", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_094", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_095", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_096", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_097", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_098", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_099", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_100", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "response_being_judged": "The study examined neural networks for image recognition. Results showed 95% accuracy. Robustness to adversarial examples remains unclear.", "ground_truth_label": "score_4", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_101", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_102", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_103", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_104", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_105", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_106", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_107", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_108", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_109", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_110", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_111", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_112", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_113", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_114", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_115", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_116", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_117", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_118", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_119", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_120", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_121", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Rate coherence 1-5. One number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "Coherence score 1 to 5, respond with number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_122", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "Assign coherence rating 1-5. Single digit response.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_123", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "Rate coherence 1-5. One number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_124", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Coherence score 1 to 5, respond with number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "How coherent is this? Score: 1=poor 5=excellent. Number only.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}
+{"pair_id": "cohe_125", "task_type": "coherence", "source_benchmark": "SummEval", "prompt_a": "Assign coherence rating 1-5. Single digit response.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "prompt_b": "Coherence: 1 (incoherent) to 5 (very coherent). Reply with number.\n\nText: Dogs have four legs. They bark. Cats have tails. Birds fly.", "response_being_judged": "Dogs have four legs. They bark. Cats have tails. Birds fly.", "ground_truth_label": "score_5", "semantic_equivalence_score": 1.0}