Causal discovery on WCHSU-Pain original (test)
80ACCTLVD
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| TLVD2026.02 | 80 | 86 | 45.3 | |
| OpenAI-deepresearch2026.02 | 55 | 25.3 | 20 | |
| Autogen2026.02 | 35 | 8.9 | 8.4 | |
| Qwen-deepresearch2026.02 | 30 | 50 | 10.5 | |
| MiniMax2026.02 | 30 | 7.1 | 5.3 | |
| Gemini-deepresearch2026.02 | 25 | 33.8 | 28.4 | |
| Multi-Agent Debate2026.02 | 20 | 0 | 0 | |
| GPT52026.02 | 15 | 9.5 | 9.5 | |
| Doubao-deepresearch2026.02 | 15 | 0 | 0 | |
| CAMEL2026.02 | 0 | 0 | 0 | |
| Multi-Agent-Majority2026.02 | 0 | 0 | 0 |