Retrieval-Augmented Question Answering on DeepSearch Bamboogle
72Success Rate (SR)Synapse
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SynapseTaskExecutor=Gemini-3.0-flash2026.04 | 72 | 44.6 | |
| ActiveContextTaskExecutor=Gemini-3.0-flash2026.04 | 66 | 5.6 | |
| Full ContextTaskExecutor=Gemini-3.0-flash2026.04 | 62 | 31.6 | |
| SynapseTaskExecutor=Gemini-2.5-flash2026.04 | 62 | 52 | |
| ActiveContextTaskExecutor=Gemini-2.5-flash2026.04 | 61.9 | 5.6 | |
| Training-Free GRPOTaskExecutor=Gemini-3.0-flash2026.04 | 60 | 33.4 | |
| SynapseTaskExecutor=gpt-4o-mini, Category=Baselines & Retrieval-Based Memory2026.04 | 60 | 30.9 | |
| SynapseTaskExecutor=gpt-4o2026.04 | 58 | 27.4 | |
| Gemini-3.0-flashTaskExecutor=gpt-4o-mini, Category=Zero-Shot Curators2026.04 | 56 | 6.3 | |
| Gemini-2.5-flashTaskExecutor=gpt-4o-mini, Category=Zero-Shot Curators2026.04 | 52 | 6.1 | |
| Full ContextTaskExecutor=Gemini-2.5-flash2026.04 | 46 | 28.8 | |
| Full ContextTaskExecutor=gpt-4o2026.04 | 46 | 28.8 | |
| ActiveContextTaskExecutor=gpt-4o2026.04 | 44 | 6.4 | |
| gpt-4oTaskExecutor=gpt-4o-mini, Category=Zero-Shot Curators2026.04 | 44 | 9.2 | |
| Qwen2.5-7B-InstructTaskExecutor=gpt-4o-mini, Category=Zero-Shot Curators2026.04 | 44 | 8.7 | |
| Training-Free GRPOTaskExecutor=Gemini-2.5-flash2026.04 | 42 | 63.5 | |
| Full ContextTaskExecutor=gpt-4o-mini, Category=Baselines & Retrieval-Based Memory2026.04 | 40 | 30 | |
| Mem0TaskExecutor=gpt-4o-mini, Category=Baselines & Retrieval-Based Memory2026.04 | 40 | 45.4 | |
| gpt-4o-miniTaskExecutor=gpt-4o-mini, Category=Zero-Shot Curators2026.04 | 40 | 8.9 | |
| AWMTaskExecutor=gpt-4o-mini, Category=Baselines & Retrieval-Based Memory2026.04 | 38 | 32.1 | |
| Training-Free GRPOTaskExecutor=gpt-4o2026.04 | 34 | 31.9 | |
| Training-Free GRPOTaskExecutor=gpt-4o-mini, Category=Baselines & Retrieval-Based Memory2026.04 | 34 | 35 | |
| ContextCuratorTaskExecutor=gpt-4o-mini, Category=RL-Trained Curator2026.04 | 32 | 7.3 |