Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Scientific Question Answering on ARC Challenge (Acc., F1, FCR, Time)

63.5Accuracy

Qwen2.5-1.5B

14.6227.314052.69May 12, 2026May 14, 2026May 16, 2026May 18, 2026May 20, 2026May 22, 2026May 25, 2026
Updated 8d ago

Evaluation Results

MethodLinks
2026.05
63.564.399.99.3
2026.05
60.761.299.74.7
2026.05
47.347.499.221.2
2026.05
42.743.699.95
2026.05
25.92398.216.7
2026.05
25.7---
2026.05
25.5---
2026.05
25.3---
2026.05
23.6---
2026.05
16.518.184.688