Chemistry Reasoning on ChemCoT (test)
69.8Und ScoreLLMZERO
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| LLMZEROType=Adaptive, Backbone=Qwen3-4B, Iterations=162026.06 | 69.8 | 33.3 | 18.5 | 40.5 | |
| Random searchType=Static, Backbone=Qwen3-4B, Iterations=162026.06 | 66.4 | 23 | 28.7 | 39.4 | |
| Skill-based LLM agentType=Adaptive, Backbone=Qwen3-4B, Iterations=162026.06 | 61.8 | 21.4 | 27.7 | 37 | |
| Grid searchType=Static, Backbone=Qwen3-4B, Iterations=162026.06 | 61.2 | 23.4 | 20.9 | 35.2 | |
| PractitionerType=Static, Backbone=Qwen3-4B, Iterations=162026.06 | 61.1 | 23 | 14.9 | 33 | |
| Qwen3-4BType=–, Backbone=Qwen3-4B, Iterations=162026.06 | 40.5 | 8 | 2.3 | 16.9 |