Causal Discovery on WCHSU-Cancer (n=12) original (test)
0.833ACCTLVD
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| TLVD2026.02 | 0.833 | 0.9 | 0.415 | |
| MiniMax2026.02 | 0.467 | 0.6 | 0.231 | |
| OpenAI-deepresearch2026.02 | 0.433 | 0.56 | 0.215 | |
| Autogen2026.02 | 0.3 | 0 | 0 | |
| Gemini-deepresearch2026.02 | 0.267 | 0.286 | 0.154 | |
| CAMEL2026.02 | 0.267 | 0 | 0 | |
| Multi-Agent Debate2026.02 | 0.234 | 0 | 0 | |
| Qwen-deepresearch2026.02 | 0.2 | 0.5 | 0.077 | |
| GPT52026.02 | 0.134 | 0.1 | 0.031 | |
| Multi-Agent-Majority2026.02 | 0.134 | 0 | 0 | |
| Doubao-deepresearch2026.02 | 0 | 0.15 | 0.038 |