E-Commerce Disputes Verdict on CyberJurors 1.0 (test)
72.92AccuracyCyberJurors
Evaluation results
| Method | Links | ||
|---|---|---|---|
| CyberJurorsType=Court Simulators2026.05 | 72.92 | 0.9098 | |
| AgentCourtType=Court Simulators2026.05 | 66.73 | 0.7051 | |
| ChatEvalType=Court Simulators2026.05 | 65.89 | 0.721 | |
| Gemini-3-ProType=Closed MLLM2026.05 | 63.54 | 0.1114 | |
| GPT-5.2-ChatType=Closed LLM2026.05 | 63.44 | 0.6834 | |
| DeepSeek-V3Type=Closed LLM2026.05 | 60.8 | 0.7008 | |
| Qwen-PlusType=Closed LLM2026.05 | 60.55 | 0.6116 | |
| Claude-Opus-4.5Type=Closed MLLM2026.05 | 59.1 | 0.7623 | |
| Doubao-Seed-1.6Type=Closed MLLM2026.05 | 56.26 | 0.3159 | |
| Grok-4Type=Closed MLLM2026.05 | 54.36 | 0.574 | |
| Gemini-2.5-Flash-LiteType=Closed MLLM2026.05 | 52.92 | 0.5705 | |
| Llama-2-13BType=Open LLM2026.05 | 50.42 | 0.7675 | |
| GLM-4Type=Closed LLM2026.05 | 50.17 | 0.382 | |
| Dolphin3.0-R1-24BType=Open LLM2026.05 | 49.29 | 0.7469 | |
| Qwen3-VL-235BType=Open MLLM2026.05 | 48.43 | 0.1331 | |
| GPT-5.2Type=Closed MLLM2026.05 | 48.33 | 0.2311 | |
| Llama-3.2-90B-VisionType=Open MLLM2026.05 | 40.9 | — |