Local Causal Attribution on ZebraLogic
348.71AUPCAttriCoT-2x
Evaluation Results
| Method | Links | |
|---|---|---|
| AttriCoT-2xModel=DS-Qwen-14B2026.06 | 348.71 | |
| AttriCoTModel=DS-Qwen-14B2026.06 | 347.59 | |
| TA-KL (no log)Model=DS-Qwen-14B2026.06 | 335.28 | |
| TA-KL (no log)Model=Qwen3-8B2026.06 | 292.42 | |
| AttriCoT-2xModel=Qwen3-8B2026.06 | 291.94 | |
| AttriCoTModel=Qwen3-8B2026.06 | 291.73 | |
| AttriCoT-2xModel=DS-Llama-8B2026.06 | 277.66 | |
| AttriCoTModel=DS-Llama-8B2026.06 | 277.11 | |
| TA-KL (no log)Model=DS-Llama-8B2026.06 | 273.68 | |
| Prompted (best same-size)Model=DS-Llama-8B2026.06 | 211.19 | |
| AttriCoT-2xModel=DS-Qwen3-8B2026.06 | 174.75 | |
| AttriCoTModel=DS-Qwen3-8B2026.06 | 174.17 | |
| TA-KL (no log)Model=DS-Qwen3-8B2026.06 | 173.69 | |
| Prompted (best same-size)Model=DS-Qwen-14B2026.06 | 167.67 | |
| Prompted-ICL (GPT-OSS)Model=DS-Qwen-14B2026.06 | 157.67 | |
| Prompted (best same-size)Model=Qwen3-8B2026.06 | 156.24 | |
| Prompted-ICL (GPT-OSS)Model=Qwen3-8B2026.06 | 135.95 | |
| Prompted-ICL (GPT-OSS)Model=DS-Llama-8B2026.06 | 126.37 | |
| Prompted (best same-size)Model=DS-Qwen3-8B2026.06 | 102.9 | |
| Prompted-ICL (GPT-OSS)Model=DS-Qwen3-8B2026.06 | 85.55 |