Citation Generation on CITEVERIFIER Citation Generation
100JSON AccClaude 4
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Claude 42026.02 | 100 | 28,800 | 21.84 | 1.93 | |
| Qwen-32026.02 | 99.38 | 28,546 | 23.52 | 1.99 | |
| Grok 42026.02 | 99.03 | 28,627 | 79.98 | 1.88 | |
| GPT-52026.02 | 97.9 | 28,366 | 50.92 | 2.36 | |
| DeepSeek2026.02 | 97.67 | 27,973 | 14.23 | 1.65 | |
| Llama 42026.02 | 97.33 | 27,925 | 45.84 | 2.36 | |
| GLM-4.52026.02 | 97.22 | 27,835 | 21.25 | 1.94 | |
| Gemini2026.02 | 95.8 | 27,102 | 59.47 | 2.34 | |
| ERNIE2026.02 | 95.06 | 27,332 | 71.9 | 2.15 | |
| Phi-42026.02 | 93.35 | 27,164 | 87.47 | 1.6 | |
| Kimi2026.02 | 89.09 | 24,955 | 41.86 | 2.44 | |
| Hunyuan2026.02 | 62.9 | 16,094 | 94.93 | 1.29 | |
| Seed2026.02 | 48.75 | 11,090 | 78.64 | 2.74 |