List your company

Science Question Answering on ARC Challenge (Acc. and Final Gap)

78.4AccuracyPioneer Agent
51.8858.76565.6572.535Apr 10, 2026

Evaluation results

MethodLinks
Pioneer AgentModel=Llama 3.2-3B, System Strategy=Agent, Checkpoint Stage=Best2026.0478.422.6
Pioneer AgentModel=Llama 3.2-3B, System Strategy=Agent, Checkpoint Stage=C12026.0474.222.6
Naive BaselineModel=Llama 3.2-3B, System Strategy=Naive, Checkpoint Stage=C42026.0455.822.6
Naive BaselineModel=Llama 3.2-3B, System Strategy=Naive, Checkpoint Stage=C12026.0452.922.6