Veracity Prediction on Factify-2
78Macro F1Mistral-7BQwenVL
Evaluation results
| Method | Links | |
|---|---|---|
| Mistral-7BQwenVLBackbone=Mistral-7B, Justification Generator=QwenVL, Protocol=QLoRA2025.05 | 78 | |
| RoBERTa-LQwenVLBackbone=RoBERTa-Large, Justification Generator=QwenVL, Protocol=Fine-tuning2025.05 | 76 | |
| XLNet-LQwenVLBackbone=XLNet-Large, Justification Generator=QwenVL, Protocol=Fine-tuning2025.05 | 76 | |
| RoBERTa-LIdefics3Backbone=RoBERTa-Large, Justification Generator=Idefics3, Protocol=Fine-tuning2025.05 | 75 | |
| Llama-8BQwenVLBackbone=Llama-8B, Justification Generator=QwenVL, Protocol=QLoRA2025.05 | 75 | |
| Llama-8BIdefics3Backbone=Llama-8B, Justification Generator=Idefics3, Protocol=QLoRA2025.05 | 75 | |
| Mistral-7BIdefics3Backbone=Mistral-7B, Justification Generator=Idefics3, Protocol=QLoRA2025.05 | 75 | |
| XLNet-LIdefics3Backbone=XLNet-Large, Justification Generator=Idefics3, Protocol=Fine-tuning2025.05 | 74 | |
| Llama-8BPaligemmaBackbone=Llama-8B, Justification Generator=Paligemma, Protocol=QLoRA2025.05 | 73 | |
| XLNet-LPaligemmaBackbone=XLNet-Large, Justification Generator=Paligemma, Protocol=Fine-tuning2025.05 | 72 | |
| Mistral-7BPaligemmaBackbone=Mistral-7B, Justification Generator=Paligemma, Protocol=QLoRA2025.05 | 72 | |
| RoBERTa-LPaligemmaBackbone=RoBERTa-Large, Justification Generator=Paligemma, Protocol=Fine-tuning2025.05 | 71 | |
| Du et al., 2023Protocol=Baseline2025.05 | 57 |