2 papers
cs.LG2026
ID Balancing: Stable Training of Extremely Sparse MoE via PID-Based Load Control
Peng Jin, Zihan Qiu, Zekun Wang +8
Scaling Large Language Models (LLMs) via Mixture-of-Experts (MoE) enables massive parameter growth with nearly constant per-token computation. However, further scaling the paramete…
cs.AI2026
MeanFlowAdvantage: Stable Reward Fine-Tuning for Few-Step Average-Velocity Generators
Haocheng Tang, Tianchi Xie, Xingqiao Lin
MeanFlow enables efficient few-step generation by predicting interval-average velocities, but this representation creates a mismatch for reward fine-tuning: existing advantage-base…