List your company

LLM Agent Performance on AgentBoard BabyAI Compositional Stream

55.4Success Rate (1)MEMPROBE
36.78441.61746.4551.283Jun 1, 2026

Evaluation results

MethodLinks
MEMPROBE2026.0655.465.753.669
ReMem2026.0651.868.662.575.7
ReAct2026.065063.65063.6
ExpRAG2026.0637.553.158.973.3