List your company

Generalization Evaluation on AM-Thinking Distilled-math&code v1 (Unseen)

49.55Average Unseen ScoreEqual
21.449228.744636.0443.3354Jul 1, 2026

Evaluation results

MethodLinks
EqualBackbone=Qwen2.5-7B, Samples=800K, Mixture Source=tulu-3-sft-mixture2026.0749.55
CAUSALMIX-ABackbone=Qwen2.5-7B, Samples=800K, Mixture Source=tulu-3-sft-mixture2026.0749.09
DMOBackbone=Qwen2.5-7B, Samples=800K, Mixture Source=tulu-3-sft-mixture2026.0748.98
RegMixBackbone=Qwen2.5-7B, Samples=800K, Mixture Source=tulu-3-sft-mixture2026.0748.12
ODMBackbone=Qwen2.5-7B, Samples=800K, Mixture Source=tulu-3-sft-mixture2026.0748
CAUSALMIX-SBackbone=Qwen2.5-7B, Samples=800K, Mixture Source=tulu-3-sft-mixture2026.0747.98
Qwen2.5-7B-InstructModel=Qwen2.5-7B-Instruct2026.0746.94
GridBackbone=Qwen2.5-7B, Samples=800K, Mixture Source=tulu-3-sft-mixture2026.0746.37
DoReMiBackbone=Qwen2.5-7B, Samples=800K, Mixture Source=tulu-3-sft-mixture2026.0746.19
Llama-3.1-Tulu-3-8B-SFTModel=Llama-3.1-Tulu-3-8B-SFT2026.0741.46
Qwen2.5-0.5B-InstructModel=Qwen2.5-0.5B-Instruct2026.0728.6
GridBackbone=Qwen2.5-0.5B, Samples=400K, Mixture Source=tulu-3-sft-mixture2026.0727.08
CAUSALMIX-SBackbone=Qwen2.5-0.5B, Samples=400K, Mixture Source=tulu-3-sft-mixture2026.0726.93
DMOBackbone=Qwen2.5-0.5B, Samples=800K, Mixture Source=tulu-3-sft-mixture2026.0726.49
RegMixBackbone=Qwen2.5-0.5B, Samples=100K, Mixture Source=tulu-3-sft-mixture2026.0726.24
DMOBackbone=Qwen2.5-0.5B, Samples=100K, Mixture Source=tulu-3-sft-mixture2026.0726.08
RegMixBackbone=Qwen2.5-0.5B, Samples=400K, Mixture Source=tulu-3-sft-mixture2026.0726.07
CAUSALMIX-SBackbone=Qwen2.5-0.5B, Samples=100K, Mixture Source=tulu-3-sft-mixture2026.0725.9
DMOBackbone=Qwen2.5-0.5B, Samples=400K, Mixture Source=tulu-3-sft-mixture2026.0725.84
DoReMiBackbone=Qwen2.5-0.5B, Samples=400K, Mixture Source=tulu-3-sft-mixture2026.0725.69
EqualBackbone=Qwen2.5-0.5B, Samples=800K, Mixture Source=tulu-3-sft-mixture2026.0725.64
ODMBackbone=Qwen2.5-0.5B, Samples=400K, Mixture Source=tulu-3-sft-mixture2026.0725.51
ODMBackbone=Qwen2.5-0.5B, Samples=800K, Mixture Source=tulu-3-sft-mixture2026.0725.36
CAUSALMIX-SBackbone=Qwen2.5-0.5B, Samples=800K, Mixture Source=tulu-3-sft-mixture2026.0725.04
CAUSALMIX-ABackbone=Qwen2.5-0.5B, Samples=800K, Mixture Source=tulu-3-sft-mixture2026.0725.02
DoReMiBackbone=Qwen2.5-0.5B, Samples=800K, Mixture Source=tulu-3-sft-mixture2026.0724.88
EqualBackbone=Qwen2.5-0.5B, Samples=400K, Mixture Source=tulu-3-sft-mixture2026.0724.71
EqualBackbone=Qwen2.5-0.5B, Samples=100K, Mixture Source=tulu-3-sft-mixture2026.0724.43
CAUSALMIX-ABackbone=Qwen2.5-0.5B, Samples=400K, Mixture Source=tulu-3-sft-mixture2026.0724.26
DoReMiBackbone=Qwen2.5-0.5B, Samples=100K, Mixture Source=tulu-3-sft-mixture2026.0724.12
ODMBackbone=Qwen2.5-0.5B, Samples=100K, Mixture Source=tulu-3-sft-mixture2026.0723.59
GridBackbone=Qwen2.5-0.5B, Samples=100K, Mixture Source=tulu-3-sft-mixture2026.0723.45
CAUSALMIX-ABackbone=Qwen2.5-0.5B, Samples=100K, Mixture Source=tulu-3-sft-mixture2026.0723.42
RegMixBackbone=Qwen2.5-0.5B, Samples=800K, Mixture Source=tulu-3-sft-mixture2026.0722.82
GridBackbone=Qwen2.5-0.5B, Samples=800K, Mixture Source=tulu-3-sft-mixture2026.0722.53