List your company

Mathematical Reasoning on MATH500 (pass@k metrics)

91.3Pass@1PieceHint-Nemotron-1.5B
37.53251.49165.4579.409Oct 11, 2025Nov 11, 2025Dec 12, 2025Jan 13, 2026Feb 13, 2026Mar 16, 2026Apr 17, 2026

Evaluation results

MethodLinks
PieceHint-Nemotron-1.5BModel Name=Nemotron-1.5B, Model Size=1.5B, Method Type=PieceHint2026.0491.3—
PieceHint-Qwen3-1.7BModel Name=Qwen3-1.7B, Model Size=1.7B, Method Type=PieceHint2026.0489.8—
Nemotron-1.5BModel Name=Nemotron-1.5B, Model Size=1.5B, Method Type=Baseline2026.0488.5—
PieceHint-DeepSeek-1.5BModel Name=DeepSeek-R1-Distill-1.5B, Model Size=1.5B, Method Type=PieceHint2026.0487.7—
DeepSeek-R1-Distill-32BModel Name=DeepSeek-R1-Distill-32B, Model Size=32B, Method Type=Baseline2026.0486—
Qwen3-4BModel Name=Qwen3-4B, Model Size=4B, Method Type=Baseline2026.0485.1—
Qwen3-1.7BModel Name=Qwen3-1.7B, Model Size=1.7B, Method Type=Baseline2026.0484.1—
OPOBackbone=Qwen2.5-Math-7B2025.1082.2—
STEERBackbone=Qwen2.5-Math-7B2025.1082.2—
DeepSeek-R1-Distill-7BModel Name=DeepSeek-R1-Distill-7B, Model Size=7B, Method Type=Baseline2026.0480.3—
GRPO w/ Entro. LossBackbone=Qwen2.5-Math-7B2025.1080—
Eurus-PRIMEBackbone=Qwen2.5-Math-7B2025.1079.8—
Entro. Adv.Backbone=Qwen2.5-Math-7B2025.1079.6—
SimpleRL-ZooBackbone=Qwen2.5-Math-7B2025.1079.2—
GRPO w/ Fork TokensBackbone=Qwen2.5-Math-7B2025.1079.2—
W-REINFORCEBackbone=Qwen2.5-Math-7B2025.1079.2—
F-DAPOBackbone=Qwen2.5-7B, Group Size (N)=82026.0279.196.6
F-CISPOBackbone=Qwen2.5-7B, Group Size (N)=82026.027997.8
GRPO w/ clip-highBackbone=Qwen2.5-Math-7B2025.1079—
KL–CovBackbone=Qwen2.5-Math-7B2025.1078.8—
CISPOBackbone=Qwen2.5-7B, Group Size (N)=82026.0278.797
DAPOBackbone=Qwen2.5-7B, Group Size (N)=82026.0278.695.2
GRPOBackbone=Qwen2.5-Math-7B2025.1078.6—
Clip–CovBackbone=Qwen2.5-Math-7B2025.1078—
CISPOBackbone=Qwen2.5-1.5B-Math, Group Size (N)=82026.0277.395.5
F-DAPOBackbone=Qwen2.5-1.5B-Math, Group Size (N)=82026.0276.897
F-CISPOBackbone=Qwen2.5-1.5B-Math, Group Size (N)=82026.0276.896.5
DAPOBackbone=Qwen2.5-1.5B-Math, Group Size (N)=82026.0276.596.2
F-GRPOBackbone=Qwen2.5-7B, Group Size (N)=82026.0276.295.1
GRPOBackbone=Qwen2.5-7B, Group Size (N)=82026.0275.892.8
GRPOBackbone=Qwen2.5-1.5B-Math, Group Size (N)=82026.0275.495.6
F-GRPOBackbone=Qwen2.5-1.5B-Math, Group Size (N)=82026.0274.796.1
DeepSeek-R1-Distill-1.5BModel Name=DeepSeek-R1-Distill-1.5B, Model Size=1.5B, Method Type=Baseline2026.0472.1—
F-DAPOBackbone=Llama-3.2-3B-Instruct, Group Size (N)=82026.0258.692
CISPOBackbone=Llama-3.2-3B-Instruct, Group Size (N)=82026.0256.989.1
F-CISPOBackbone=Llama-3.2-3B-Instruct, Group Size (N)=82026.0256.591
DAPOBackbone=Llama-3.2-3B-Instruct, Group Size (N)=82026.0255.983.8
GRPOBackbone=Llama-3.2-3B-Instruct, Group Size (N)=82026.025590.6
F-GRPOBackbone=Llama-3.2-3B-Instruct, Group Size (N)=82026.0254.192.9
Qwen2.5-Math-7BBackbone=Qwen2.5-Math-7B2025.1039.6—