Reading Comprehension on DROP (Performance and Cost)
83.49Performance ScoreFlowBank
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| FlowBankExecutor LLM=GPT-4o mini, Optimizer LLM=Qwen3-8B2026.06 | 83.49 | 1.06 | |
| ScoreFlowExecutor LLM=GPT-4o mini, Generator LLM=Qwen3-8B2026.06 | 82.1 | 1.42 | |
| MedPromptExecutor LLM=GPT-4o mini2026.06 | 81.77 | 1.75 | |
| ADASExecutor LLM=GPT-4o mini, Optimizer LLM=GPT-4o2026.06 | 81.17 | 0.98 | |
| AFlow (GPT-4o)Executor LLM=GPT-4o mini, Optimizer LLM=GPT-4o2026.06 | 80.26 | 1.02 | |
| Multi-agent DebateExecutor LLM=GPT-4o mini2026.06 | 78.65 | 1.64 | |
| GPTSwarmExecutor LLM=GPT-4o mini2026.06 | 78.37 | 0.51 | |
| Self-ConsistencyExecutor LLM=GPT-4o mini2026.06 | 78.33 | 0.74 | |
| CoTExecutor LLM=GPT-4o mini2026.06 | 77.96 | 0.11 | |
| ComplexCoTExecutor LLM=GPT-4o mini2026.06 | 77.74 | 0.33 | |
| AFlow (Qwen3-8B)Executor LLM=GPT-4o mini, Optimizer LLM=Qwen3-8B2026.06 | 77.08 | 1.05 | |
| AgentSquareExecutor LLM=GPT-4o mini, Optimizer LLM=GPT-4o2026.06 | 76.84 | 0.76 | |
| Self-RefineExecutor LLM=GPT-4o mini2026.06 | 76.71 | 0.34 | |
| MaASExecutor LLM=GPT-4o mini2026.06 | 76.64 | 1.64 | |
| MultiPersonaExecutor LLM=GPT-4o mini2026.06 | 76.13 | 0.29 | |
| IOExecutor LLM=GPT-4o mini2026.06 | 75.49 | 0.08 |