tracking shuffled objects seven objects on BBH (test)
92.8AccuracyTextGrad
Evaluation Results
| Method | Links | |
|---|---|---|
| TextGradTest Engine=Phi-3.5-Mini-Instruct2026.05 | 92.8 | |
| TextRegTest Engine=Phi-3.5-Mini-Instruct2026.05 | 92.2 | |
| CoTTest Engine=Phi-3.5-Mini-Instruct2026.05 | 89.1 | |
| EvoPrompt(GA)-OPTS(TS)Task-solving LLM=GPT-4o mini, Number of trials=32025.03 | 88.33 | |
| REVOLVETest Engine=Phi-3.5-Mini-Instruct2026.05 | 87.7 | |
| EvoPrompt(DE)-OPTS(TS)Task-solving LLM=GPT-4o mini2025.03 | 81.83 | |
| EvoPrompt(DE)Task-solving LLM=GPT-4o mini2025.03 | 80.67 | |
| TextRegTest Engine=Llama-3.1-8B-Instruct2026.05 | 76.6 | |
| EvoPrompt(GA)Task-solving LLM=GPT-4o mini, Number of trials=32025.03 | 75.67 | |
| TextGradTest Engine=Llama-3.1-8B-Instruct2026.05 | 66.7 | |
| REVOLVETest Engine=Llama-3.1-8B-Instruct2026.05 | 65.4 | |
| CoTTest Engine=Llama-3.1-8B-Instruct2026.05 | 48.5 | |
| TextRegTest Engine=Llama-3-8B-Instruct2026.05 | 48.3 | |
| REVOLVETest Engine=Llama-3-8B-Instruct2026.05 | 41.8 | |
| REVOLVETest Engine=Qwen2-7B-Instruct2026.05 | 40.2 | |
| TextRegTest Engine=Qwen2-7B-Instruct2026.05 | 39.1 | |
| TextGradTest Engine=Llama-3-8B-Instruct2026.05 | 38 | |
| CoTTest Engine=Llama-3-8B-Instruct2026.05 | 36.5 | |
| CoTTest Engine=Qwen2-7B-Instruct2026.05 | 33.9 | |
| TextGradTest Engine=Qwen2-7B-Instruct2026.05 | 33.1 |