A dynamic two-population co-evolutionary algorithm (CHEA), which balances feasibility, convergence and diversity at different stages by dynamically adjusting the number of offspring of the two populations by dynamically adjusting the number of offspring of the two populations.
Mengduo Yu, Sheng-Wei Wang, Keda Chen et al.· Cluster Computing· 0 citations
By prioritizing operator over aggressive operand precision, the ATOM-synthesized data outperforms rigorous baselines (e.g., +3.1% gain over LIMA), suggesting that operator diversity matters more than operand precision.
Jia-Xiang Liu, Chenhao Yuan, Shunnan Xu et al.· 0 citations
Diffusion large language models (dLLMs) are a promising alternative to autoregressive generation. However, reasoning-oriented post-training for dLLMs remains challenging. Supervised fine-tuning (SFT) for dLLMs requires dense but often off-policy masked states, while reinforcement learning (RL) relies on sparse rewards...