Selective Entropy Regularization Boosts Large Reasoning Models
SIREN, a selective entropy regularization method, boosted the Qwen2.5‑Math‑7B model by 6.6 points on the majority‑at‑k metric for AIME24/25 benchmarks. Read more: getnews.me/selective-entropy-regula... #siren #largereasoningmodels #entropyregularization
0
0
0
0