Chat on AlpacaEval Length Controlled (test)
27.86AlpLC ScoreG-Zero
Evaluation Results
| Method | Links | |
|---|---|---|
| G-ZeroBackbone=Llama-3.1-8B-Instruct, Training Method=G-Zero, Training Round=22026.05 | 27.86 | |
| base modelBackbone=Llama-3.1-8B-Instruct, Training Method=Base2026.05 | 24.12 | |
| G-ZeroBackbone=Llama-3.1-8B-Instruct, Training Method=G-Zero, Training Round=12026.05 | 23.88 | |
| R-ZeroBackbone=Llama-3.1-8B-Instruct, Training Method=R-Zero2026.05 | 21.74 | |
| G-ZeroBackbone=Qwen3-8B-Base, Training Method=G-Zero, Training Round=12026.05 | 9.07 | |
| base modelBackbone=Qwen3-8B-Base, Training Method=Base2026.05 | 8.94 | |
| G-ZeroBackbone=Qwen3-8B-Base, Training Method=G-Zero, Training Round=22026.05 | 8.47 | |
| R-ZeroBackbone=Qwen3-8B-Base, Training Method=R-Zero2026.05 | 8.04 |