Short-form generation on Short-form generation ID
67PRRHBO
Evaluation Results
| Method | Links | |
|---|---|---|
| HBOModel=Gemma-2-9B2026.04 | 67 | |
| HBOModel=Llama 3.1-8B2026.04 | 66 | |
| Uhead (1 layer)Model Backbone=Gemma-2-9B2026.04 | 66 | |
| Uhead (2 layers)Model Backbone=Gemma-2-9B2026.04 | 65 | |
| SATRMD-MSPModel=Gemma-2-9B2026.04 | 64 | |
| HUQ-SATRMDModel=Gemma-2-9B2026.04 | 64 | |
| MSPModel=Gemma-2-9B2026.04 | 64 | |
| SATRMD-MSPModel Backbone=Gemma-2-9B2026.04 | 64 | |
| HUQ-SATRMDModel Backbone=Gemma-2-9B2026.04 | 64 | |
| MSPModel Backbone=Gemma-2-9B2026.04 | 64 | |
| SAPLMA (mid)Model=Llama 3.1-8B2026.04 | 63 | |
| SAPLMA (middle)Model Backbone=Llama 3.1-8B2026.04 | 63 | |
| Uhead (1 layer)Model Backbone=Llama 3.1-8B2026.04 | 63 | |
| Uhead (2 layers)Model Backbone=Llama 3.1-8B2026.04 | 63 | |
| HUQ-SATMDModel=Gemma-2-9B2026.04 | 61 | |
| SATMDModel Backbone=Gemma-2-9B2026.04 | 61 | |
| HUQ-SATMDModel Backbone=Gemma-2-9B2026.04 | 61 | |
| SATRMD-MSPModel=Llama 3.1-8B2026.04 | 60 | |
| HUQ-SATRMDModel=Llama 3.1-8B2026.04 | 60 | |
| SATRMD-MSPModel Backbone=Llama 3.1-8B2026.04 | 60 | |
| HUQ-SATRMDModel Backbone=Llama 3.1-8B2026.04 | 60 | |
| HUQ-SATMDModel=Llama 3.1-8B2026.04 | 58 | |
| SAPLMA (mid)Model=Gemma-2-9B2026.04 | 58 | |
| HUQ-SATMDModel Backbone=Llama 3.1-8B2026.04 | 58 | |
| SAPLMA (middle)Model Backbone=Gemma-2-9B2026.04 | 58 | |
| MSPModel=Llama 3.1-8B2026.04 | 57 | |
| MSPModel Backbone=Llama 3.1-8B2026.04 | 57 | |
| SAPLMA (top)Model Backbone=Gemma-2-9B2026.04 | 56 | |
| SATMD-MSPModel=Gemma-2-9B2026.04 | 54 | |
| SAPLMA (top)Model Backbone=Llama 3.1-8B2026.04 | 54 | |
| SATMD-MSPModel Backbone=Gemma-2-9B2026.04 | 54 | |
| SATRMDModel Backbone=Gemma-2-9B2026.04 | 53 | |
| SATRMDModel Backbone=Llama 3.1-8B2026.04 | 47 | |
| Lookback-lensModel Backbone=Gemma-2-9B2026.04 | 34 | |
| SATMD-MSPModel=Llama 3.1-8B2026.04 | 33 | |
| SATMD-MSPModel Backbone=Llama 3.1-8B2026.04 | 33 | |
| SATMDModel Backbone=Llama 3.1-8B2026.04 | 30 | |
| Lookback-lensModel Backbone=Llama 3.1-8B2026.04 | 27 |