List your company

Mathematical Reasoning on OMNI-MATH (Overall Accuracy)

39.55Overall AccuracyTEPO
2.557212.161121.76531.3689Apr 14, 2026

Evaluation results

MethodLinks
TEPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0439.55
GRPO/DAPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0438.42
TEPOBackbone=Qwen3-14B, Method Variant=TEPO2026.0435.11
DeepSeek-R1-Distill-Qwen-7BBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0433.86
Qwen3-14B w. GSPOBackbone=Qwen3-14B, Method Variant=GSPO2026.0433.54
TEPOBackbone=DeepSeek-R1-Distill-Llama-8B2026.0433.11
Qwen3-14B w. Entropy-based TermBackbone=Qwen3-14B, Method Variant=Entropy-based Term2026.0433.04
Qwen3-14B w. GRPO/DAPOBackbone=Qwen3-14B, Method Variant=GRPO/DAPO2026.0433.01
GRPO/DAPOBackbone=DeepSeek-R1-Distill-Llama-8B2026.0432.97
Qwen3-14B w. CLIP-CovBackbone=Qwen3-14B, Method Variant=CLIP-Cov2026.0432.94
Qwen3-14B w. KL-CovBackbone=Qwen3-14B, Method Variant=KL-Cov2026.0432.87
Qwen3-14B w. GPGBackbone=Qwen3-14B, Method Variant=GPG2026.0432.22
Qwen3-14BBackbone=Qwen3-14B, Method Variant=Base2026.0430.27
TEPOBackbone=Qwen2.5-7B, Method Variant=TEPO2026.0427.17
Qwen2.5-7B w. GPGBackbone=Qwen2.5-7B, Method Variant=GPG2026.0426.28
Qwen2.5-7B w. Entropy-based TermBackbone=Qwen2.5-7B, Method Variant=Entropy-based Term2026.0426.14
Qwen2.5-7B w. CLIP-CovBackbone=Qwen2.5-7B, Method Variant=CLIP-Cov2026.0426.11
Qwen2.5-7B w. GSPOBackbone=Qwen2.5-7B, Method Variant=GSPO2026.0425.72
Qwen2.5-7B w. KL-CovBackbone=Qwen2.5-7B, Method Variant=KL-Cov2026.0425.64
DeepSeek-R1-Distill-Llama-8BBackbone=DeepSeek-R1-Distill-Llama-8B2026.0425.5
Qwen2.5-7B w. GRPO/DAPOBackbone=Qwen2.5-7B, Method Variant=GRPO/DAPO2026.0425
Qwen2.5-7BBackbone=Qwen2.5-7B, Method Variant=Base2026.0413.73
TEPOBackbone=Mistral-7B-Instruct-v0.22026.044.59
GRPO/DAPOBackbone=Mistral-7B-Instruct-v0.22026.044.27
Mistral-7B-Instruct-v0.2Backbone=Mistral-7B-Instruct-v0.22026.043.98