ContextPilot: Teaching Agents for Proactive Context Management via Fine-grained RL
Open paper
Benchmark
Model
Rank
Results
agents-on-gaia-text-only-103-validation-tasks
ContextPilot
#5
Accuracy: 57.93
deep-research-agents-on-browsecomp-plus
ContextPilot-14B-RL
#13
Accuracy: 55.5
Rank counts only results with a code link.