Collection by
Mar 15
Mar 15
Collections
Discover the best community collections!
Collections including paper arxiv:2202.01169
Collection by
Mar 15
Mar 15
-
Mixture-of-Supernets: Improving Weight-Sharing Supernet Training with Architecture-Routed Mixture-of-Experts
Paper • 2306.04845 • Published • 4 -
Patch-level Routing in Mixture-of-Experts is Provably Sample-efficient for Convolutional Neural Networks
Paper • 2306.04073 • Published • 2 -
Branch-Train-MiX: Mixing Expert LLMs into a Mixture-of-Experts LLM
Paper • 2403.07816 • Published • 39 -
Unified Scaling Laws for Routed Language Models
Paper • 2202.01169 • Published • 2
Collection by
Mar 15
Mar 15
-
Turn Waste into Worth: Rectifying Top-k Router of MoE
Paper • 2402.12399 • Published • 2 -
CompeteSMoE -- Effective Training of Sparse Mixture of Experts via Competition
Paper • 2402.02526 • Published • 3 -
Buffer Overflow in Mixture of Experts
Paper • 2402.05526 • Published • 8 -
OpenMoE: An Early Effort on Open Mixture-of-Experts Language Models
Paper • 2402.01739 • Published • 26
Collection by
Jul 16
Jul 16
-
Robust Mixture-of-Expert Training for Convolutional Neural Networks
Paper • 2308.10110 • Published • 2 -
Experts Weights Averaging: A New General Training Scheme for Vision Transformers
Paper • 2308.06093 • Published • 2 -
ConstitutionalExperts: Training a Mixture of Principle-based Prompts
Paper • 2403.04894 • Published • 2 -
Mixture-of-LoRAs: An Efficient Multitask Tuning for Large Language Models
Paper • 2403.03432 • Published • 1
Collection by
Apr 3
Apr 3
-
Adaptive sequential Monte Carlo by means of mixture of experts
Paper • 1108.2836 • Published • 2 -
Convergence Rates for Mixture-of-Experts
Paper • 1110.2058 • Published • 2 -
Multi-view Contrastive Learning for Entity Typing over Knowledge Graphs
Paper • 2310.12008 • Published • 2 -
Enhancing NeRF akin to Enhancing LLMs: Generalizable NeRF Transformer with Mixture-of-View-Experts
Paper • 2308.11793 • Published • 2
Collection by
Aug 23
Aug 23
-
Non-asymptotic oracle inequalities for the Lasso in high-dimensional mixture of experts
Paper • 2009.10622 • Published • 1 -
MoE-LLaVA: Mixture of Experts for Large Vision-Language Models
Paper • 2401.15947 • Published • 49 -
MoE-Mamba: Efficient Selective State Space Models with Mixture of Experts
Paper • 2401.04081 • Published • 71 -
MoE-Infinity: Activation-Aware Expert Offloading for Efficient MoE Serving
Paper • 2401.14361 • Published • 2
Collection by
Oct 9
Oct 9
-
Measuring the Effects of Data Parallelism on Neural Network Training
Paper • 1811.03600 • Published • 2 -
Adafactor: Adaptive Learning Rates with Sublinear Memory Cost
Paper • 1804.04235 • Published • 2 -
EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
Paper • 1905.11946 • Published • 3 -
Yi: Open Foundation Models by 01.AI
Paper • 2403.04652 • Published • 62
Collection by
Feb 27
Feb 27
-
Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer
Paper • 1701.06538 • Published • 4 -
Sparse Networks from Scratch: Faster Training without Losing Performance
Paper • 1907.04840 • Published • 3 -
ZeRO: Memory Optimizations Toward Training Trillion Parameter Models
Paper • 1910.02054 • Published • 4 -
A Mixture of h-1 Heads is Better than h Heads
Paper • 2005.06537 • Published • 2
Collection by
8 days ago
8 days ago
-
Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer
Paper • 1701.06538 • Published • 4 -
Sparse Networks from Scratch: Faster Training without Losing Performance
Paper • 1907.04840 • Published • 3 -
ZeRO: Memory Optimizations Toward Training Trillion Parameter Models
Paper • 1910.02054 • Published • 4 -
A Mixture of h-1 Heads is Better than h Heads
Paper • 2005.06537 • Published • 2
Collection by
Jul 9
Jul 9
-
QMoE: Practical Sub-1-Bit Compression of Trillion-Parameter Models
Paper • 2310.16795 • Published • 26 -
Pre-gated MoE: An Algorithm-System Co-Design for Fast and Scalable Mixture-of-Expert Inference
Paper • 2308.12066 • Published • 4 -
Towards MoE Deployment: Mitigating Inefficiencies in Mixture-of-Expert (MoE) Inference
Paper • 2303.06182 • Published • 1 -
EvoMoE: An Evolutional Mixture-of-Experts Training Framework via Dense-To-Sparse Gate
Paper • 2112.14397 • Published • 1