Summarization on UniSumEval
66CompletenessJury-on-Demand
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| Jury-on-DemandJury Configuration=Jury-on-Demand2025.12 | 66 | 0.42 | 62 | — | |
| GPT-OSS 120BModel Variant=GPT-OSS 120B2025.12 | 66 | — | 59 | — | |
| GPT-OSS 20BModel Variant=GPT-OSS 20B2025.12 | 62 | — | 61 | — | |
| Gemini 2.5 FlashModel Variant=Gemini 2.5 Flash2025.12 | 59 | — | 61 | — | |
| Gemma 3Model Variant=Gemma 32025.12 | 55 | — | 55 | — | |
| Gemini 2.0 FlashModel Variant=Gemini 2.0 Flash2025.12 | 54 | — | 57 | — | |
| Claude 3.7Model Variant=Claude 3.72025.12 | 52 | — | 61 | — | |
| Gemini 2.5 ProModel Variant=Gemini 2.5 Pro2025.12 | 52 | — | 64 | — | |
| DeepSeek R1Model Variant=DeepSeek R12025.12 | 41 | — | 24 | — | |
| LLAMA 3.2Model Variant=LLAMA 3.22025.12 | 29 | — | 10 | — | |
| Phi 4Model Variant=Phi 42025.12 | 29 | — | 31 | — | |
| Static JuryJury Configuration=Average-All2025.12 | — | 0.41 | — | — | |
| Static JuryJury Configuration=Average-TopK2025.12 | — | 0.41 | — | — | |
| Static JuryJury Configuration=Weighted-Regression2025.12 | — | 0.4 | — | — | |
| Static JuryJury Configuration=Weighted-Tau2025.12 | — | 0.42 | — | — |