Causal Variable Identification on Arithmetic
88.2F1 (X)GPT-5
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| GPT-5Setting=Decompositional Framework2025.05 | 88.2 | 86.5 | 82.1 | 85.6 | |
| GPT-o4Setting=Decompositional Framework, Model Variant=GPT-o4-mini-high2025.05 | 84.9 | 82.8 | 82.3 | 73.4 | |
| Llama4-MSetting=Decompositional Framework, Model Variant=Llama-4-Maverick-17B-128E2025.05 | 81.6 | 79.8 | 80.3 | 86 | |
| Llama4-SSetting=Decompositional Framework, Model Variant=Llama-4-Scout-17B2025.05 | 80.3 | 78.6 | 79.1 | 85.1 | |
| DeepSeekSetting=Decompositional Framework, Model Variant=DeepSeek-VL2025.05 | 78.3 | 76.1 | 65.7 | 74.9 | |
| Gemini2.5Setting=Decompositional Framework, Model Variant=Gemini2.5-Pro2025.05 | 76.9 | 74.5 | 62.8 | 73.5 | |
| Qwen3Setting=Decompositional Framework, Model Variant=Qwen3-VL-235B-A22B Thinking2025.05 | 75.7 | 73.8 | 63.6 | 71.9 |