Retrieval-Augmented Question Answering on DeepSearch PopQA
64Success RateActiveContext
Evaluation results
| Method | Links | ||
|---|---|---|---|
| ActiveContextTaskExecutor=Gemini-3.0-flash2026.04 | 64 | 7.4 | |
| Full ContextTaskExecutor=Gemini-3.0-flash2026.04 | 62 | 36.4 | |
| SynapseTaskExecutor=Gemini-3.0-flash2026.04 | 54 | 100.5 | |
| SynapseTaskExecutor=Gemini-2.5-flash2026.04 | 48 | 76.4 | |
| Training-Free GRPOTaskExecutor=Gemini-3.0-flash2026.04 | 44 | 47.2 | |
| ActiveContextTaskExecutor=Gemini-2.5-flash2026.04 | 40.9 | 7.3 | |
| SynapseTaskExecutor=gpt-4o2026.04 | 40 | 47.8 | |
| Gemini-3.0-flashTaskExecutor=gpt-4o-mini, Category=Zero-Shot Curators2026.04 | 40 | 8 | |
| SynapseTaskExecutor=gpt-4o-mini, Category=Baselines & Retrieval-Based Memory2026.04 | 38 | 50.5 | |
| ContextCuratorTaskExecutor=gpt-4o-mini, Category=RL-Trained Curator2026.04 | 38 | 7.6 | |
| Training-Free GRPOTaskExecutor=gpt-4o2026.04 | 36 | 31 | |
| Full ContextTaskExecutor=gpt-4o-mini, Category=Baselines & Retrieval-Based Memory2026.04 | 34 | 33.8 | |
| AWMTaskExecutor=gpt-4o-mini, Category=Baselines & Retrieval-Based Memory2026.04 | 34 | 34.8 | |
| gpt-4oTaskExecutor=gpt-4o-mini, Category=Zero-Shot Curators2026.04 | 34 | 12 | |
| Full ContextTaskExecutor=Gemini-2.5-flash2026.04 | 32 | 35.1 | |
| Full ContextTaskExecutor=gpt-4o2026.04 | 32 | 35.1 | |
| Mem0TaskExecutor=gpt-4o-mini, Category=Baselines & Retrieval-Based Memory2026.04 | 32 | 47.9 | |
| Training-Free GRPOTaskExecutor=gpt-4o-mini, Category=Baselines & Retrieval-Based Memory2026.04 | 32 | 36.2 | |
| gpt-4o-miniTaskExecutor=gpt-4o-mini, Category=Zero-Shot Curators2026.04 | 32 | 10.8 | |
| Gemini-2.5-flashTaskExecutor=gpt-4o-mini, Category=Zero-Shot Curators2026.04 | 30 | 8.2 | |
| Qwen2.5-7B-InstructTaskExecutor=gpt-4o-mini, Category=Zero-Shot Curators2026.04 | 28 | 9.9 | |
| Training-Free GRPOTaskExecutor=Gemini-2.5-flash2026.04 | 26 | 75.9 | |
| ActiveContextTaskExecutor=gpt-4o2026.04 | 26 | 7.7 |