List your company
ResearchFormerly WizWand

First Integral Prediction on First Integral Discovery Normal (test)

74.7AccuracyFISolver-Base
10.42827.11443.860.486May 20, 2026
Updated 5mo ago

Evaluation results

MethodLinks
FISolver-Basebeam size (bs)=70, Training Strategy=RL, RL Training Data=HardSplit (HS), RL Reward Scheme=LD-based Reward (LD)2026.0574.7
FISolver-Basebeam size (bs)=70, Training Strategy=RL, RL Training Data=NormalSplit (NS), RL Reward Scheme=LD-based Reward (LD)2026.0574.2
Mathematica2026.0574
FISolver-Basebeam size (bs)=70, Training Strategy=SFT-only2026.0572.7
FISolver-Basebeam size (bs)=70, Training Strategy=RL, RL Training Data=HardSplit (HS), RL Reward Scheme=Binary Reward (Bin)2026.0571.1
FISolver-Basebeam size (bs)=70, Training Strategy=RL, RL Training Data=NormalSplit (NS), RL Reward Scheme=Binary Reward (Bin)2026.0549.5
FISolver-Basebeam size (bs)=5, Training Strategy=RL, RL Training Data=NormalSplit (NS), RL Reward Scheme=LD-based Reward (LD)2026.0549
FISolver-Basebeam size (bs)=5, Training Strategy=RL, RL Training Data=NormalSplit (NS), RL Reward Scheme=Binary Reward (Bin)2026.0545.1
FISolver-Basebeam size (bs)=5, Training Strategy=SFT-only2026.0543.2
FISolver-Basebeam size (bs)=5, Training Strategy=RL, RL Training Data=HardSplit (HS), RL Reward Scheme=LD-based Reward (LD)2026.0541.8
FISolver-Basebeam size (bs)=5, Training Strategy=RL, RL Training Data=HardSplit (HS), RL Reward Scheme=Binary Reward (Bin)2026.0537
Qwen2.5-Math-7Bbeam size (bs)=52026.0536.6
Qwen2.5-Math-1.5Bbeam size (bs)=52026.0512.9