Critical Error Detection on SynCED-EnDe 2025 (test)
95MCCGPT-OSS-20B
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| GPT-OSS-20BFine-tuning=true, Inference Mode=zero-shot, Decision Rule=majority voting2026.02 | 95 | 95 | 97 | |
| GPT-4o-miniFine-tuning=true, Inference Mode=zero-shot, Decision Rule=majority voting2026.02 | 94 | 94 | 96 | |
| LLaMA-3.1-8BFine-tuning=true, Inference Mode=zero-shot, Decision Rule=majority voting2026.02 | 92 | 96 | 96 |