Argument Quality Evaluation on WebisArgQuality 20
41.08F1 ScoreQwen3-4B-ArgumentOnly-Target-SFT
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen3-4B-ArgumentOnly-Target-SFTModel Backbone=Qwen3-4B, Finetuning Strategy=Pre-adaptive Finetuning, SFT Data Source=ArgumentOnly2026.03 | 41.08 | |
| Qwen3-4B-Arguinas-Target-SFTModel Backbone=Qwen3-4B, Finetuning Strategy=Pre-adaptive Finetuning, SFT Data Source=Arguinas2026.03 | 41.01 | |
| Qwen3-4B-Target-SFTModel Backbone=Qwen3-4B, Finetuning Strategy=Pre-adaptive Finetuning, SFT Data Source=Target2026.03 | 39.37 | |
| Qwen3-8B-EntailmentBank-Target-SFTModel Backbone=Qwen3-8B, Finetuning Strategy=Pre-adaptive Finetuning, SFT Data Source=EntailmentBank2026.03 | 38.96 | |
| Qwen3-4B-AAAC-Target-SFTModel Backbone=Qwen3-4B, Finetuning Strategy=Pre-adaptive Finetuning, SFT Data Source=AAAC2026.03 | 38.58 | |
| Qwen3-4B-EntailmentBank-Target-SFTModel Backbone=Qwen3-4B, Finetuning Strategy=Pre-adaptive Finetuning, SFT Data Source=EntailmentBank2026.03 | 38 | |
| Qwen3-8B-Arguinas-Target-SFTModel Backbone=Qwen3-8B, Finetuning Strategy=Pre-adaptive Finetuning, SFT Data Source=Arguinas2026.03 | 37.77 | |
| Qwen3-8B-AAAC-Target-SFTModel Backbone=Qwen3-8B, Finetuning Strategy=Pre-adaptive Finetuning, SFT Data Source=AAAC2026.03 | 37.67 | |
| Qwen3-4B-Instruct-Arguinas-SFTModel Backbone=Qwen3-4B-Instruct, Finetuning Strategy=Continued Finetuning, SFT Data Source=Arguinas2026.03 | 36.43 | |
| Qwen3-8B-ArgumentOnly-Target-SFTModel Backbone=Qwen3-8B, Finetuning Strategy=Pre-adaptive Finetuning, SFT Data Source=ArgumentOnly2026.03 | 35.27 | |
| Qwen3-8B-Target-SFTModel Backbone=Qwen3-8B, Finetuning Strategy=Pre-adaptive Finetuning, SFT Data Source=Target2026.03 | 34.97 | |
| Qwen3-4B-InstructModel Backbone=Qwen3-4B-Instruct, Finetuning Strategy=Continued Finetuning, SFT Data Source=Baseline2026.03 | 34.68 | |
| Qwen2.5-7B-Instruct-Arguinas-SFTModel Backbone=Qwen2.5-7B-Instruct, Finetuning Strategy=Continued Finetuning, SFT Data Source=Arguinas2026.03 | 33.82 | |
| Qwen2.5-7B-InstructModel Backbone=Qwen2.5-7B-Instruct, Finetuning Strategy=Continued Finetuning, SFT Data Source=Baseline2026.03 | 22.35 |