Long-form generation on Long-form generation ID
0.38PRRHUQ-SATRMD
Evaluation Results
| Method | Links | |
|---|---|---|
| HUQ-SATRMDModel=Llama 3.1-8B2026.04 | 0.38 | |
| HUQ-SATRMDModel Backbone=Llama 3.1-8B2026.04 | 0.38 | |
| SATMD-MSPModel=Gemma-2-9B2026.04 | 0.34 | |
| SATMD-MSPModel Backbone=Gemma-2-9B2026.04 | 0.34 | |
| SAPLMA (mid)Model=Llama 3.1-8B2026.04 | 0.32 | |
| SAPLMA (mid)Model=Gemma-2-9B2026.04 | 0.32 | |
| SAPLMA (middle)Model Backbone=Llama 3.1-8B2026.04 | 0.32 | |
| SAPLMA (middle)Model Backbone=Gemma-2-9B2026.04 | 0.32 | |
| Uhead (1 layer)Model Backbone=Llama 3.1-8B2026.04 | 0.31 | |
| SATMD-MSPModel=Llama 3.1-8B2026.04 | 0.3 | |
| SATMD-MSPModel Backbone=Llama 3.1-8B2026.04 | 0.3 | |
| Uhead (1 layer)Model Backbone=Gemma-2-9B2026.04 | 0.28 | |
| Uhead (2 layers)Model Backbone=Llama 3.1-8B2026.04 | 0.27 | |
| HUQ-SATMDModel=Gemma-2-9B2026.04 | 0.26 | |
| HUQ-SATMDModel Backbone=Gemma-2-9B2026.04 | 0.26 | |
| Uhead (2 layers)Model Backbone=Gemma-2-9B2026.04 | 0.23 | |
| SAPLMA (top)Model Backbone=Gemma-2-9B2026.04 | 0.22 | |
| Lookback-lensModel Backbone=Gemma-2-9B2026.04 | 0.22 | |
| SATRMD-MSPModel=Gemma-2-9B2026.04 | 0.21 | |
| SATRMD-MSPModel Backbone=Gemma-2-9B2026.04 | 0.21 | |
| HBOModel=Llama 3.1-8B2026.04 | 0.2 | |
| HUQ-SATRMDModel=Gemma-2-9B2026.04 | 0.2 | |
| SAPLMA (top)Model Backbone=Llama 3.1-8B2026.04 | 0.2 | |
| SATRMDModel Backbone=Gemma-2-9B2026.04 | 0.2 | |
| HUQ-SATRMDModel Backbone=Gemma-2-9B2026.04 | 0.2 | |
| SATRMD-MSPModel=Llama 3.1-8B2026.04 | 0.17 | |
| SATRMD-MSPModel Backbone=Llama 3.1-8B2026.04 | 0.17 | |
| SATMDModel Backbone=Gemma-2-9B2026.04 | 0.17 | |
| SATRMDModel Backbone=Llama 3.1-8B2026.04 | 0.16 | |
| SATMDModel Backbone=Llama 3.1-8B2026.04 | 0.14 | |
| Lookback-lensModel Backbone=Llama 3.1-8B2026.04 | 0.14 | |
| HBOModel=Gemma-2-9B2026.04 | 0.03 | |
| HUQ-SATMDModel=Llama 3.1-8B2026.04 | 0.02 | |
| HUQ-SATMDModel Backbone=Llama 3.1-8B2026.04 | 0.02 | |
| MSPModel=Llama 3.1-8B2026.04 | -0.01 | |
| MSPModel Backbone=Llama 3.1-8B2026.04 | -0.01 | |
| MSPModel=Gemma-2-9B2026.04 | -0.1 | |
| MSPModel Backbone=Gemma-2-9B2026.04 | -0.1 |