LSH-MoE: Communication-efficient MoE Training via Locality-Sensitive Hashing

Feb-15-2026, 06:42:26 GMT–Neural Information Processing Systems

GPUs remains a significant challenge.

large language model, machine learning, natural language, (20 more...)

Neural Information Processing Systems

Feb-15-2026, 06:42:26 GMT

Conferences PDF

Country:
- Asia
  - China
    - Beijing > Beijing (0.04)
    - Shandong Province > Qingdao (0.04)
  - Middle East > Jordan (0.04)
- North America > Canada
  - Quebec > Montreal (0.04)

Genre:
- Research Report > Experimental Study (0.93)

Industry:
- Information Technology (0.46)

Technology:
- Information Technology > Artificial Intelligence
  - Machine Learning > Neural Networks
    - Deep Learning (1.00)
  - Natural Language > Large Language Model (0.93)
  - Representation & Reasoning (0.67)
  - Vision (1.00)

Duplicate Docs Excel Report

Title
LSH-MoE: Communication-efficient MoE Training via Locality-Sensitive Hashing

Similar Docs Excel Report more

Title	Similarity	Source
None found