Contradiction Detection on RevCI (test)
19.01FNRIMPACT-P
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| IMPACT-PCategory=Our Proposed2026.05 | 19.01 | 16.13 | |
| IMPACT-OACategory=Our Proposed2026.05 | 23.9 | 22.87 | |
| CourtEvalCategory=Generic multi-agent2026.05 | 25.2 | 25.9 | |
| ChatEvalCategory=Generic multi-agent2026.05 | 25.8 | 26.4 | |
| LLaMA-4 MaverickCategory=Single-agent (CoT)2026.05 | 26.39 | 46.8 | |
| MADCategory=Generic multi-agent2026.05 | 26.9 | 27.6 | |
| Self-RefineCategory=Generic multi-agent2026.05 | 27.5 | 29.5 | |
| DebateCategory=Generic multi-agent2026.05 | 28.7 | 28.2 | |
| GPT-5.2Category=Single-agent (CoT)2026.05 | 29.35 | 30.12 | |
| LLaMA-3-8B-InstructCategory=Single-agent (CoT)2026.05 | 35.7 | 39.48 | |
| Gemini-3 FlashCategory=Single-agent (CoT)2026.05 | 37.41 | 23.69 | |
| TIDECategory=Our Proposed2026.05 | 37.71 | 30.48 | |
| ContraSciViewCategory=Single-agent (CoT)2026.05 | 39.2 | 33.6 | |
| Qwen-2.5-7B-InstructCategory=Single-agent (CoT)2026.05 | 41.27 | 45.34 |