CEA assembles per-token low-rank residual updates via dense affinities over hyper-adapter-generated components to improve all-in-one image restoration on spatially non-uniform degradations.
Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity.Journal of Machine Learning Research, 23 (120):1–39
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 8roles
background 1polarities
background 1representative citing papers
Joint few-step distillation and step-specific structural pruning turns a video diffusion model into a compact Mixture-of-Models that cuts 24% extra FLOPs per step and reaches 30× speedup on Wan-14B.
MCWC aligns permutation-symmetric blocks across layers to enable sequential prediction and residual entropy coding, improving rate-accuracy tradeoffs versus quantization and prior codecs on language and vision models.
BitsMoE uses SVD decomposition and activation-aware ILP bit allocation to quantize MoE LLMs at ultra-low bits with reduced accuracy degradation compared to GPTQ.
PRiSE-EEG is a prior-guided EEG foundation model that allocates shared and specialized experts across depth using CKA-derived sigmoid mappings and reports strong cross-paradigm results on 12 benchmarks.
A new MoE training method integrates expert-level losses and partial online updates to improve forecasting accuracy and efficiency over standard statistical and neural models.
A shared global expert pool in MoE improves validation loss over per-layer experts and allows sublinear expert-parameter growth with depth.
MoP combines specialized parallelisms and a new optimizer step for memory-efficient MoE training, delivering 4.7x-8.2x higher per-GPU throughput than FSDP2 while supporting 1M-token contexts on 12 8x H200 nodes.
citing papers explorer
-
Continuous Expert Assembly: Instance-Conditioned Low-Rank Residuals for All-in-One Image Restoration
CEA assembles per-token low-rank residual updates via dense affinities over hyper-adapter-generated components to improve all-in-one image restoration on spatially non-uniform degradations.
-
Dynamic-in-Few-Step: Unifying Dynamic Computation and Few-Step Distillation for Efficient Video Generation
Joint few-step distillation and step-specific structural pruning turns a video diffusion model into a compact Mixture-of-Models that cuts 24% extra FLOPs per step and reaches 30× speedup on Wan-14B.
-
Motion-Compensated Weight Compression
MCWC aligns permutation-symmetric blocks across layers to enable sequential prediction and residual entropy coding, improving rate-accuracy tradeoffs versus quantization and prior codecs on language and vision models.
-
BitsMoE: Efficient Spectral Energy-Guided Bit Allocation for MoE LLM Quantization
BitsMoE uses SVD decomposition and activation-aware ILP bit allocation to quantize MoE LLMs at ultra-low bits with reduced accuracy degradation compared to GPTQ.
-
PRiSE-EEG: A Prior-Guided Foundation Model with Depth-Stratified Experts for Cross-Paradigm EEG Representation Learning
PRiSE-EEG is a prior-guided EEG foundation model that allocates shared and specialized experts across depth using CKA-derived sigmoid mappings and reports strong cross-paradigm results on 12 benchmarks.
-
Fast Training of Mixture-of-Experts for Time Series Forecasting via Expert Loss Integration
A new MoE training method integrates expert-level losses and partial online updates to improve forecasting accuracy and efficiency over standard statistical and neural models.
-
UniPool: A Globally Shared Expert Pool for Mixture-of-Experts
A shared global expert pool in MoE improves validation loss over per-layer experts and allows sublinear expert-parameter growth with depth.
-
Mixture-of-Parallelisms: Towards Memory-Efficient Training Stack for Mixture-of-Experts Models
MoP combines specialized parallelisms and a new optimizer step for memory-efficient MoE training, delivering 4.7x-8.2x higher per-GPU throughput than FSDP2 while supporting 1M-token contexts on 12 8x H200 nodes.