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

Multiple Choice Question Answering on MMLU (Detailed Subject Accuracy)

67.1STEM Accuracy

Original

22.27633.91345.5557.187Jun 3, 2024Sep 23, 2024Jan 13, 2025May 6, 2025Aug 26, 2025Dec 16, 2025Apr 8, 2026
Updated 9d ago

Evaluation Results

MethodLinks
2026.04
67.1-84.1-68.6-78.7---
2026.04
62.7-80.4-64.1-75.3---
2026.04
60.2-78.5-62.3-73.4---
2026.04
59.5-78-60.4-73---
2026.04
54.6-72.9-58.4-67.9---
2026.04
54.1-74.9-58-70.2---
2026.04
52.3-67.2-53.9-64.2---
2026.04
51.3-69.7-57.8-64.2---
2026.02
50.52-65.58-55.98-----
2026.02
50.49-65.52-55.9-----
2026.02
50.02-64.64-56.13-----
2026.02
49.92-65.42-56.49-----
2026.02
49.83-65-55.73-----
2024.06
48.736.268.56254.243.364.861.858.650.2
2024.06
4842.468.565.858.452.265.6626056.8
2024.06
47.541.86865.255.844.364.661.858.754.3
2024.06
46.942.368.1645547.965.559.959.254.8
2024.06
4439.562.759.553.142.460.959.35549.5
2024.06
43.833.562.554.552.243.360.651.654.645.6
2024.06
43.63562.155.950.545.460.454.953.847.7
2024.06
43.23962.356.852.145.560.556.854.449
2026.04
42-57.7-45.7-45.7---
2026.04
36.1-47.9-40.4-49.5---
2026.04
35.8-48.3-39.7-50---
2026.04
35.3-47.6-39.7-50---
2026.04
34.9-47.3-38.7-46.9---
2026.04
34.1-47.4-38.3-47.4---
2026.04
33.9-46.4-38.1-47.7---
2026.04
33.7-47-39.2-46.6---
2026.04
33.6-41.6-34.3-43.1---
2026.04
32.9-42.5-35.9-45.6---
2026.04
24.4-25.4-27.3-27.9---
2026.04
24-25.8-22.7-23.4---
2026.02
--------60.6-
2026.02
--------60.9-
2026.02
--------60.7-
2026.02
--------60.66-
2026.02
--------62.05-