Ranking on BEIR selected subset v1.0.0 (test)
84.7TREC-COVIDLISTT5-3B
Evaluation Results
| Method | Links | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| LISTT5-3BModel type=Listwise, Backbone=T5-3B, r=22024.02 | 84.7 | 71.8 | 69.1 | 37.7 | 33.8 | 53.2 | 57.8 | 33.6 | 46.2 | 77 | 70.5 | 53 | |
| RankZephyr-7bModel type=Listwise, Backbone=Zephyr-7b2024.02 | 84 | 73.9 | 70.9 | 36.7 | 31.8 | 52.6 | 54.3 | 33.8 | 44.6 | 74.9 | 72.4 | 51.6 | |
| listwiseLLM=Qwen2.5-7B2026.02 | 82.34 | — | — | 39.05 | 33.63 | 49.89 | 54.38 | 31.84 | 43.24 | 71.07 | — | 50.68 | |
| setwiseLLM=Mistral-7B2026.02 | 81.6 | — | — | 36.15 | 29.19 | 45.2 | 48.13 | 28.79 | 39.53 | 62.6 | — | 46.4 | |
| listwiseLLM=LLaMa3.1-8B2026.02 | 81.12 | — | — | 36.9 | 31.07 | 48.33 | 52.01 | 27.44 | 41.89 | 67.62 | — | 48.3 | |
| setwiseLLM=Qwen2.5-7B2026.02 | 80.99 | — | — | 37.86 | 29.99 | 47.88 | 52.1 | 28.83 | 40.74 | 68 | — | 48.3 | |
| RankVicuna-7bModel type=Listwise, Backbone=Vicuna-7b2024.02 | 80.5 | 68.9 | 66.1 | 33.2 | 34.2 | 46.9 | 48.9 | 33 | 44.4 | 70.8 | 67.5 | 49 | |
| RankSteerLLM=Mistral-7B2026.02 | 80.19 | — | — | 36.23 | 28.77 | 44.41 | 51.23 | 22.54 | 39.65 | 56.64 | — | 44.96 | |
| RankSteerBackbone=Mistral-7B2026.02 | 80.19 | — | — | 36.23 | 28.77 | 44.41 | 51.23 | 22.54 | 39.65 | 56.64 | — | 44.96 | |
| DuoT5-baseModel type=Pairwise, Backbone=T5-base, Reranked by=MonoT5 on top 50 passages2024.02 | 80.1 | 71.4 | 67.4 | 35 | 31.4 | 49.1 | 49.6 | 31.8 | 43.9 | 69.6 | 69.4 | 48.8 | |
| listwiseLLM=Mistral-7B2026.02 | 80.06 | — | — | 37.13 | 32.68 | 50.92 | 53.59 | 35.21 | 44.25 | 65.84 | — | 49.96 | |
| pairwiseLLM=Qwen2.5-7B2026.02 | 79.89 | — | — | 37.96 | 31.11 | 49.68 | 54.15 | 30.38 | 41.34 | 69.68 | — | 49.27 | |
| pairwiseLLM=Mistral-7B2026.02 | 79.66 | — | — | 35.41 | 30.36 | 48.54 | 49.81 | 30.99 | 33.54 | 66.55 | — | 46.96 | |
| RG-S(0,4)Backbone=LLaMa3.1-8B2026.02 | 79.02 | — | — | 36.46 | 26.37 | 44.51 | 50.39 | 22.05 | 36.66 | 62.41 | — | 44.73 | |
| RG-S(0,4)Backbone=Mistral-7B2026.02 | 78.92 | — | — | 35.12 | 26.79 | 46.78 | 48.56 | 24.68 | 36.72 | 63.24 | — | 45.1 | |
| RankSteerLLM=LLaMa3.1-8B2026.02 | 78.63 | — | — | 37.32 | 27.79 | 44.62 | 52.07 | 24.45 | 37.52 | 64.98 | — | 45.88 | |
| RankSteerBackbone=LLaMa3.1-8B2026.02 | 78.63 | — | — | 37.32 | 27.79 | 44.62 | 52.07 | 24.45 | 37.52 | 64.98 | — | 45.88 | |
| pairwiseLLM=LLaMa3.1-8B2026.02 | 78.35 | — | — | 36.58 | 28.91 | 47 | 51.15 | 25.78 | 38 | 70.32 | — | 47.01 | |
| LISTT5-baseModel type=Listwise, Backbone=T5-base, r=22024.02 | 78.3 | 71.8 | 68.1 | 35.6 | 33.5 | 48.5 | 52.1 | 33.4 | 43.7 | 74.1 | 70 | 49.9 | |
| RG-S(0,4)Backbone=Qwen2.5-7B2026.02 | 78.14 | — | — | 35.99 | 26.61 | 44.9 | 47.56 | 23.33 | 36.18 | 62.61 | — | 44.41 | |
| RG-YN_roleBackbone=LLaMa3.1-8B2026.02 | 77.92 | — | — | 37.15 | 26.61 | 44.56 | 53.21 | 22.27 | 35.74 | 66.06 | — | 45.44 | |
| setwiseLLM=LLaMa3.1-8B2026.02 | 77.89 | — | — | 34.94 | 29.46 | 46.14 | 49.65 | 24.7 | 37.74 | 68.08 | — | 46.08 | |
| PAGC-YGBackbone=Mistral-7B2026.02 | 77.75 | — | — | 33.38 | 25.86 | 39.49 | 48.94 | 22.12 | 33.07 | 44.76 | — | 40.67 | |
| RankSteerLLM=Qwen2.5-7B2026.02 | 76.78 | — | — | 36.97 | 27.93 | 38.07 | 43.38 | 23.56 | 37.49 | 41.33 | — | 40.69 | |
| RankSteerBackbone=Qwen2.5-7B2026.02 | 76.78 | — | — | 36.97 | 27.93 | 38.07 | 43.38 | 23.56 | 37.49 | 41.33 | — | 40.69 | |
| RankGPTModel type=Listwise, Backbone=GPT-3.5-turbo-03012024.02 | 76.7 | 65.8 | 62.9 | 35.6 | 32.1 | 48.9 | 50.6 | 36.2 | 44.5 | 70.4 | 64.4 | 49.4 | |
| PAGC-YGBackbone=LLaMa3.1-8B2026.02 | 75.34 | — | — | 35.96 | 25.86 | 45.03 | 51.99 | 24.24 | 34.2 | 64.85 | — | 44.68 | |
| PAGC-YGBackbone=Qwen2.5-7B2026.02 | 73.95 | — | — | 35.52 | 27.61 | 37.72 | 44.88 | 24.13 | 35.25 | 46.32 | — | 40.67 | |
| RG-YNBackbone=Mistral-7B2026.02 | 73.15 | — | — | 33.35 | 22.43 | 39.63 | 47.6 | 19.76 | 27.91 | 51.43 | — | 39.41 | |
| RG-YNBackbone=LLaMa3.1-8B2026.02 | 72.61 | — | — | 35.63 | 24.82 | 36.65 | 50.46 | 22.64 | 30.93 | 56.77 | — | 41.31 | |
| RG-YN_roleBackbone=Qwen2.5-7B2026.02 | 72.12 | — | — | 31.83 | 22.49 | 30.48 | 36.91 | 19.85 | 29.09 | 23.98 | — | 33.34 | |
| RG-YN_roleBackbone=Mistral-7B2026.02 | 71.36 | — | — | 34.01 | 22.68 | 41.92 | 48.84 | 19.98 | 29.54 | 53.04 | — | 40.17 | |
| QGBackbone=LLaMa3.1-8B2026.02 | 69.25 | — | — | 35.06 | 10.15 | 42.21 | 45.08 | 22.05 | 28.17 | 60.41 | — | 39.05 | |
| RG-YNBackbone=Qwen2.5-7B2026.02 | 67.8 | — | — | 31.06 | 23.05 | 30.4 | 34.84 | 20.25 | 27.33 | 23.95 | — | 32.34 | |
| QGBackbone=Mistral-7B2026.02 | 66.21 | — | — | 30.19 | 10.1 | 30.48 | 35.06 | 22.77 | 30.75 | 48.16 | — | 34.22 | |
| QGBackbone=Qwen2.5-7B2026.02 | 64.63 | — | — | 28.45 | 9.22 | 25.33 | 36.17 | 21.9 | 25.3 | 34.99 | — | 30.75 | |
| BM252026.02 | 59.47 | — | — | 32.18 | 33.04 | 39.52 | 40.7 | 44.22 | 31.8 | 67.89 | — | 43.6 | |
| BM252026.02 | 59.47 | — | — | 32.18 | 33.04 | 39.52 | 40.7 | 44.22 | 31.8 | 67.89 | — | 43.6 |