3 papers
cs.LG2026
UniPool: A Globally Shared Expert Pool for Mixture-of-Experts
Minbin Huang, Han Shi, Chuanyang Zheng +5
Modern Mixture-of-Experts (MoE) architectures allocate expert capacity through a rigid per-layer rule: each transformer layer owns a separate expert set. This convention couples de…
cs.CV2025
Answer-Consistent Chain-of-thought Reinforcement Learning For Multi-modal Large Langauge Models
Minbin Huang, Runhui Huang, Chuanyang Zheng +4
Recent advances in large language models (LLMs) have demonstrated that reinforcement learning with verifiable rewards (RLVR) can significantly enhance reasoning abilities by direct…
cs.LG2025
Logits-Based Finetuning
Jingyao Li, Senqiao Yang, Sitong Wu +4
In recent years, developing compact and efficient large language models (LLMs) has emerged as a thriving area of research. Traditional Supervised Fine-Tuning (SFT), which relies on…