List your company

Mathematical Reasoning on AMC 23 (Avg@16)

69.1Avg@16HEAPA
27.08437.99248.959.808Jan 30, 2026

Evaluation results

MethodLinks
HEAPABackbone=Qwen3-8B2026.0169.1
GRPO (w. PS)Backbone=Qwen3-8B2026.0168.6
GRPOBackbone=Qwen3-8B2026.0168
GRPOBackbone=Qwen3-4B2026.0158.9
GRPO (w. PS)Backbone=Qwen3-4B2026.0158.7
HEAPABackbone=Qwen3-4B2026.0158.6
GRPO (w. PS)Backbone=Qwen3-1.7B2026.0140.6
HEAPABackbone=Qwen3-1.7B2026.0140.3
GRPOBackbone=Qwen3-1.7B2026.0139.8
HEAPABackbone=Qwen3-0.6B2026.0131.9
GRPO (w. PS)Backbone=Qwen3-0.6B2026.0130.6
GRPOBackbone=Qwen3-0.6B2026.0128.7