Binary Classification on Conspiratorial Content Dataset
90AccuracyTelConGBERT
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| TelConGBERTEvaluation Protocol=fine-tuned2025.12 | 90 | 82 | 81 | 81 | |
| roberta-largeEvaluation Protocol=fine-tuned2025.12 | 85 | 87 | 86 | 87 | |
| google/gemini-2.0-flashEvaluation Protocol=zero-shot/few-shot2025.12 | 84 | 86 | 85 | 85 | |
| meta-llama/Llama-3.2-3BEvaluation Protocol=zero-shot/few-shot2025.12 | 83 | 84 | 83 | 83 | |
| deepset/gbert-baseEvaluation Protocol=fine-tuned2025.12 | 83 | 81 | 82 | 82 | |
| GPT-4Evaluation Protocol=zero-shot/few-shot2025.12 | 83 | 80 | 84 | 82 | |
| roberta-baseEvaluation Protocol=fine-tuned2025.12 | 80 | 83 | 81 | 82 | |
| microsoft/deberta-baseEvaluation Protocol=fine-tuned2025.12 | 80 | 80 | 80 | 80 | |
| GPT-3.5Evaluation Protocol=zero-shot/few-shot2025.12 | 78 | 74 | 78 | 76 | |
| bert-large-uncasedEvaluation Protocol=fine-tuned2025.12 | 77 | 77 | 77 | 77 | |
| distilbert-base-uncasedEvaluation Protocol=fine-tuned2025.12 | 76 | 79 | 78 | 78 | |
| bert-base-uncasedEvaluation Protocol=fine-tuned2025.12 | 75 | 77 | 76 | 77 | |
| Goyal et al.Evaluation Protocol=zero-shot/few-shot2025.12 | 72 | 85 | 65 | 74 | |
| Llama 2Evaluation Protocol=zero-shot/few-shot2025.12 | 66 | 57 | 60 | 60 | |
| aisingapore/SEA-LION-v1-7BEvaluation Protocol=zero-shot/few-shot2025.12 | 63 | 68 | 64 | 65 |