Question Answering on CONFLICTS
62.9Exact MatchGemma-2-27B†
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Gemma-2-27B†Params=27B, Protocol=NF, Baseline Category=Strong-Baseline (NF ≥ 30%)2026.06 | 62.9 | — | |
| Gemma-2-27B†Params=27B, Protocol=PDE, Baseline Category=Strong-Baseline (NF ≥ 30%)2026.06 | 62 | -0.9 | |
| Qwen2.5-32B†Params=32B, Protocol=PDE, Baseline Category=Strong-Baseline (NF ≥ 30%)2026.06 | 62 | 0.8 | |
| Qwen2.5-14B†Params=14B, Protocol=PDE, Baseline Category=Strong-Baseline (NF ≥ 30%)2026.06 | 61.2 | 3 | |
| Qwen2.5-32B†Params=32B, Protocol=NF, Baseline Category=Strong-Baseline (NF ≥ 30%)2026.06 | 61.2 | — | |
| Qwen3-8BParams=8B, Protocol=NF, Baseline Category=Strong-Baseline (NF ≥ 30%)2026.06 | 60.8 | — | |
| Gemma-2-9BParams=9B, Protocol=NF, Baseline Category=Strong-Baseline (NF ≥ 30%)2026.06 | 60.8 | — | |
| Qwen2.5-7BParams=7B, Protocol=NF, Baseline Category=Strong-Baseline (NF ≥ 30%)2026.06 | 59.9 | — | |
| Gemma-2-9BParams=9B, Protocol=PDE, Baseline Category=Strong-Baseline (NF ≥ 30%)2026.06 | 59.9 | -0.9 | |
| Qwen2.5-7BParams=7B, Protocol=PDE, Baseline Category=Strong-Baseline (NF ≥ 30%)2026.06 | 58.6 | -1.3 | |
| Qwen3-8BParams=8B, Protocol=PDE, Baseline Category=Strong-Baseline (NF ≥ 30%)2026.06 | 58.6 | -2.2 | |
| Qwen2.5-14B†Params=14B, Protocol=NF, Baseline Category=Strong-Baseline (NF ≥ 30%)2026.06 | 58.2 | — | |
| Llama-3.1-8BParams=8B, Protocol=PDE, Baseline Category=Weak-Baseline (NF < 30%)2026.06 | 50.2 | 36.3 | |
| Mistral-7B-v0.3Params=7B, Protocol=PDE, Baseline Category=Weak-Baseline (NF < 30%)2026.06 | 43.5 | 25.4 | |
| Mistral-7B-v0.3Params=7B, Protocol=NF, Baseline Category=Weak-Baseline (NF < 30%)2026.06 | 18.1 | — | |
| Llama-3.1-8BParams=8B, Protocol=NF, Baseline Category=Weak-Baseline (NF < 30%)2026.06 | 13.9 | — |