Inventory Management on Inventory Management (test)
103.33Const UniInvAgent
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| InvAgentReasoning effort=high, LLM=GPT-5, Variant=w/ strategy2026.02 | 103.33 | 93.37 | 388.89 | 170.25 | 64.39 | 164.05 | |
| InvAgentReasoning effort=high, LLM=GPT-5, Variant=w/ step_desc2026.02 | 76.67 | 40.06 | 169.33 | 95.04 | 191.67 | 114.55 | |
| InvAgentReasoning effort=high, LLM=GPT-5, Variant=w/o strategy2026.02 | 73.33 | 51.2 | 173.33 | 111.98 | 71.97 | 96.36 | |
| AIM-RMReasoning effort=high, LLM=GPT-5, Variant=w/ RL_log2026.02 | 28.67 | 44.88 | 195.56 | 102.07 | 134.24 | 101.08 | |
| AIM-RMReasoning effort=high, LLM=GPT-5, Variant=w/o RL_log2026.02 | 23.33 | 42.11 | 175.11 | 114.38 | 158.64 | 102.71 | |
| InvAgentReasoning effort=high, LLM=GPT-5, Variant=w/ step_desc and our_strategy2026.02 | 0 | 73.8 | 202.02 | 264.46 | 444.7 | 197.04 |