List your company

Mathematical Reasoning on AIME25 (Acc@16)

22Avg Score @16HEAPA
2.5527.60112.6517.699Jan 30, 2026

Evaluation results

MethodLinks
HEAPABackbone=Qwen3-8B2026.0122—
GRPO (w. PS)Backbone=Qwen3-8B2026.0121.9—
GRPOBackbone=Qwen3-8B2026.0121.7—
GRPO (w. PS)Backbone=Qwen3-4B2026.0119.6—
HEAPABackbone=Qwen3-4B2026.0119.3—
GRPOBackbone=Qwen3-4B2026.0118.1—
GRPOBackbone=Qwen3-1.7B2026.016.2—
GRPO (w. PS)Backbone=Qwen3-1.7B2026.016.2—
HEAPABackbone=Qwen3-1.7B2026.016.2—
HEAPABackbone=Qwen3-0.6B2026.014.3—
GRPO (w. PS)Backbone=Qwen3-0.6B2026.013.9—
GRPOBackbone=Qwen3-0.6B2026.013.3—