Composition-RL
Datasets and trained checkpoints of Composition-RL: https://github.com/XinXU-USTC/Composition-RL
Viewer • Updated • 1.32M • 14 • 1Note Compositional prompts constructed from Polaris53K
xx18/Composition-RL-EVA
Viewer • Updated • 12.8k • 30 • 1Note Evaluation datasets of Composition-RL
xx18/MATH-Composition-199K
Viewer • Updated • 199k • 20 • 1Note The training set of Composition-RL, consists of 199K compositional prompts constructed from MATH12K
xx18/Composition-RL-4B
UpdatedNote Initial Model: Qwen3-4B-Base; Training set: MATH-Composition-199K
xx18/Composition-RL-8B
Updated • 1Note Initial Model: Qwen3-8B-Base; Training set: MATH-Composition-199K
xx18/Composition-RL-14B
UpdatedNote Initial Model: Qwen3-14B-Base; Training set: MATH-Composition-199K
xx18/Composition-RL-30B-A3B
UpdatedNote Initial Model: Qwen3-30B-A3B-Base; Training set: MATH-Composition-199K
xx18/Physics-MATH-Composition-141K
Viewer • Updated • 141k • 14Note The training set of cross-domain experiments of Composition-RL, consists of 141K compositional prompts constructed from the physics subset of MegaScience and MATH12K.
xx18/Composition-RL-4B-Physics_Math
Updated • 4Note Initial Model: Qwen3-4B-Base; Training set: Physics-MATH-Composition-141K
xx18/MATH-Composition-Depth3
Viewer • Updated • 132k • 13Note Compositional prompts of Depth 3
xx18/Baseline-4B-MATH12K
UpdatedNote Initial Model: Qwen3-4B-Base; Training set: MATH12K
xx18/Composition-RL-4B-Depth1_2
UpdatedNote Initial Model: Baseline-4B-MATH12K; Training set: MATH-Composition-199K
xx18/Composition-RL-4B-Depth1_2_3
Updated • 1Note Initial Model: Composition-RL-4B-Depth1_2; Training set: MATH-Composition-Depth3