Opinion Summarization Metric Evaluation on OpinSummEval
65Aspect RelevanceBARTScore_rev->hyp
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| BARTScore_rev->hypEvaluation Level=System-level2026.02 | 65 | 76 | 77 | 46 | 1.5 | |
| SummaQA_confEvaluation Level=System-level2026.02 | 58 | 60 | 69 | 44 | 2.8 | |
| BLANC_helpEvaluation Level=System-level2026.02 | 56 | 54 | 62 | 38 | 5.5 | |
| SummaQA_fscoreEvaluation Level=System-level2026.02 | 56 | 58 | 66 | 38 | 4.5 | |
| ChatGPT-[gpt-3.5-turbo]Evaluation Level=System-level2026.02 | 56 | 62 | 56 | 62 | 3.3 | |
| FactSimEvaluation Level=System-level2026.02 | 56 | 58 | 62 | 42 | 4.3 | |
| SUPERTEvaluation Level=System-level2026.02 | 54 | 56 | 60 | 40 | 6.3 | |
| BLANC_tuneEvaluation Level=System-level2026.02 | 49 | 47 | 55 | 31 | 8.5 | |
| QAFactEvalEvaluation Level=System-level2026.02 | 45 | 47 | 51 | 27 | 9.8 | |
| G-Eval-[gpt-3.5-turbo]Evaluation Level=System-level2026.02 | 45 | 56 | 55 | 53 | 6.3 | |
| SummaC_docEvaluation Level=System-level2026.02 | 30 | 27 | 40 | 11 | 11.5 | |
| ChatGPT-[gpt-3.5-turbo]Evaluation Level=Summary-level2026.02 | 30 | 25 | 33 | 42 | 2 | |
| FactSimEvaluation Level=Summary-level2026.02 | 26 | 21 | 27 | 16 | 3.8 | |
| G-Eval-[gpt-3.5-turbo]Evaluation Level=Summary-level2026.02 | 23 | 26 | 34 | 36 | 2 | |
| BARTScore_rev->hypEvaluation Level=Summary-level2026.02 | 22 | 29 | 34 | 33 | 2.3 | |
| SummaC_sntEvaluation Level=System-level2026.02 | 19 | 16 | 22 | -4 | 12.8 | |
| BLANC_helpEvaluation Level=Summary-level2026.02 | 17 | 16 | 24 | 17 | 4.8 | |
| SUPERTEvaluation Level=Summary-level2026.02 | 17 | 14 | 18 | 12 | 6.8 | |
| BLANC_tuneEvaluation Level=Summary-level2026.02 | 14 | 10 | 19 | 9 | 8.3 | |
| SummaQA_fscoreEvaluation Level=Summary-level2026.02 | 10 | 9 | 17 | 10 | 9 | |
| SummaQA_confEvaluation Level=Summary-level2026.02 | 10 | 12 | 14 | 15 | 8.3 | |
| QAFactEvalEvaluation Level=Summary-level2026.02 | 8 | 9 | 21 | 13 | 8.3 | |
| SummaC_docEvaluation Level=Summary-level2026.02 | 6 | 1 | 17 | 5 | 11.3 | |
| G-Eval-[text-ada-001]-nEvaluation Level=System-level2026.02 | 5 | 12 | 40 | 29 | 11.8 | |
| G-Eval-[text-ada-001]-nEvaluation Level=Summary-level2026.02 | 1 | 14 | 7 | 6 | 11 | |
| SummaC_sntEvaluation Level=Summary-level2026.02 | 0 | 1 | 12 | 1 | 12.8 | |
| G-Eval-[text-ada-001]Evaluation Level=System-level2026.02 | -1 | -5 | 22 | 27 | 13.3 | |
| G-Eval-[text-ada-001]Evaluation Level=Summary-level2026.02 | -1 | -2 | 8 | 8 | 13 | |
| PPL-[PEGASUS]Evaluation Level=Summary-level2026.02 | -6 | -7 | -3 | -7 | 15 | |
| PPL-[PEGASUS]Evaluation Level=System-level2026.02 | -8 | -1 | 2 | -22 | 14.8 | |
| PPL-[GPT-2]Evaluation Level=System-level2026.02 | -10 | -8 | 0 | -24 | 16 | |
| PPL-[GPT-2]Evaluation Level=Summary-level2026.02 | -14 | -15 | -5 | -16 | 16 |