3 papers
cs.LG2026
When Non-Adversarial Noise Hurts Learning: Benchmarking Skill Self-Evolution in Agents
Tao Yu, Zhongtian Luo, Yuchen Long +19
Large language model agents increasingly improve by distilling experience into reusable skills, but errors in that experience can persist in learned rules and affect later tasks. E…
cs.AI2026
SpecFit: Not All Draft Tokens Cost the Same in Offloaded MoE Inference
Hongliang Cao, Xiuyu Li, Siqi Wang +3
Mixture-of-experts (MoE) models activate only a few experts per token, but when their expert weights are offloaded to host memory, host-to-device (H2D) expert transfers enter the c…
cs.DC2026
RailWave: Adaptive Spatial and Temporal Scheduling for Expert-Parallel Communication
Chutian Wang, Wenhao He, Jingmin Zhu +3
Irregular All-to-All communication is a major bottleneck in expert-parallel Mixture-of-Experts (MoE) models. Even with fixed expert routing and placement, uneven utilization of par…