Reading Comprehension on DROP (test) (Accuracy)
90.8AccuracyBayesFlow
Evaluation results
| Method | Links | |
|---|---|---|
| BayesFlowOptimizer=Claude-3.5-Sonnet, Executor=Claude-3.7-Sonnet2026.01 | 90.8 | |
| AFlowOptimizer=Claude-3.5-Sonnet, Executor=Claude-3.7-Sonnet2026.01 | 89.2 | |
| MaASOptimizer=Claude-3.5-Sonnet, Executor=Claude-3.7-Sonnet2026.01 | 84.2 | |
| Meta-AgentLLM Backbone=GPT-4o-mini, Evaluation Split=Zhang et al. [2025d], Run Type=single-run2026.05 | 82.7 | |
| CoTOptimizer=Claude-3.5-Sonnet, Executor=Claude-3.7-Sonnet2026.01 | 81.5 | |
| ADASOptimizer=Claude-3.5-Sonnet, Executor=Claude-3.7-Sonnet2026.01 | 81.2 | |
| AFlowLLM Backbone=GPT-4o-mini, Evaluation Split=Zhang et al. [2025d], Run Type=single-run2026.05 | 80.6 | |
| CoT-SCLLM Backbone=GPT-4o-mini, Evaluation Split=Zhang et al. [2025d], Run Type=single-run2026.05 | 78.8 | |
| BayesFlowOptimizer=Claude-3.5-Sonnet, Executor=Qwen 2.5-7B-Instruct2026.01 | 78.6 | |
| CoTLLM Backbone=GPT-4o-mini, Evaluation Split=Zhang et al. [2025d], Run Type=single-run2026.05 | 78.5 | |
| MedPromptLLM Backbone=GPT-4o-mini, Evaluation Split=Zhang et al. [2025d], Run Type=single-run2026.05 | 78 | |
| ADASLLM Backbone=GPT-4o-mini, Evaluation Split=Zhang et al. [2025d], Run Type=single-run2026.05 | 76.6 | |
| IOOptimizer=Claude-3.5-Sonnet, Executor=Claude-3.7-Sonnet2026.01 | 75.1 | |
| ADASOptimizer=Claude-3.5-Sonnet, Executor=Qwen 2.5-7B-Instruct2026.01 | 75.1 | |
| MultiPersonaLLM Backbone=GPT-4o-mini, Evaluation Split=Zhang et al. [2025d], Run Type=single-run2026.05 | 74.4 | |
| Self-RefineLLM Backbone=GPT-4o-mini, Evaluation Split=Zhang et al. [2025d], Run Type=single-run2026.05 | 70.2 | |
| AFlowOptimizer=Claude-3.5-Sonnet, Executor=Qwen 2.5-7B-Instruct2026.01 | 69.9 | |
| IOLLM Backbone=GPT-4o-mini, Evaluation Split=Zhang et al. [2025d], Run Type=single-run2026.05 | 68.3 | |
| CoT-SCOptimizer=Claude-3.5-Sonnet, Executor=Claude-3.7-Sonnet2026.01 | 49.9 | |
| MaASOptimizer=Claude-3.5-Sonnet, Executor=Qwen 2.5-7B-Instruct2026.01 | 48.7 | |
| CoTOptimizer=Claude-3.5-Sonnet, Executor=Qwen 2.5-7B-Instruct2026.01 | 32.2 | |
| IOOptimizer=Claude-3.5-Sonnet, Executor=Qwen 2.5-7B-Instruct2026.01 | 27.7 | |
| CoT-SCOptimizer=Claude-3.5-Sonnet, Executor=Qwen 2.5-7B-Instruct2026.01 | 12.3 |