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

Scientific Proposal Generation on AI Idea Bench and LiveIdeaBench 24 held-out benchmark groups 2025

3.04Novelty

EIG

2.14562.37782.612.8422May 6, 2026
Updated 27d ago

Evaluation Results

MethodLinks
2026.05
3.043.783.883.743.154.011.64
2.942.93.0632.992.934.86
2.92.793.033.073.062.795.51
2026.05
2.93.143.713.282.993.52.92
2.863.063.423.012.783.064.18
2026.05
2.713.13.53.083.083.313.15
2026.05
2.182.722.762.7122.645.74