Tech

Distributed LLM Inference via Communication-Efficient Speculative Decoding

By Joohyun Shin, Byeonghyeon Park, Keumseo Ryum, Joonhyuk Kang · arXiv cs.LG

cs.ITcs.LGeess.SPmath.IT

arXiv:2610.35957v1 Announce Type: new Abstract: Distributed speculative decoding has emerged as a promising approach for device--server collaborative large language model (LLM) inference, but its decoding speedup is often limited…

Published: October 28, 2026 (1 h ago)
Outlet: arXiv cs.LG
Section: Tech
Length here: Headline and 29-word snippet. The full article is at the source.

More from arXiv cs.LG and around the same time

A semiconductor wafer being lifted from processing equipment

A Graph-informed Kriging model for sample-efficient Bayesian optimization of wind farm layoutsarXiv:2610.35975v1 Announce Type: new Abstract: Wind farm layout optimization becomes increasingly difficult as turbine count grows, especially when high-fidelity aerodynamic models are used. Conventional Kriging-based Bayesian…arXiv cs.LG · 1 h ago · reads at arXiv cs.LG

Calibrated for Which Policy? Conformal Uncertainty for Fixed World Models Under InterventionarXiv:2610.35877v1 Announce Type: new Abstract: Changing a deployment policy can invalidate a world model's uncertainty while leaving the physical transition mechanism and predictor unchanged. For one-step prediction, if source…arXiv cs.LG · 1 h ago · reads at arXiv cs.LG

Condition-Enhanced Generative Adversarial Network for Aerodynamic Configuration Design of Commercial AircraftarXiv:2610.35963v1 Announce Type: new Abstract: Generative machine learning provides an end-to-end framework for aerodynamic design optimization by generating diverse aerodynamic configurations conditioned on target performance…arXiv cs.LG · 1 h ago · reads at arXiv cs.LG

Curriculum-Driven Deep Reinforcement Learning for Autonomous Far-Range Rocket LandingarXiv:2610.35954v1 Announce Type: new Abstract: Autonomous vertical landing of reusable launch vehicles is a critical yet formidable challenge in aerospace engineering, primarily due to highly nonlinear dynamics, stringent…arXiv cs.LG · 1 h ago · reads at arXiv cs.LG

Data-driven Discovery of Lie Symmetries and First Integrals with Latent ODEsarXiv:2610.35967v1 Announce Type: new Abstract: We present LatentLieODE, a novel neural ODE method to discover linear Lie symmetries and first integrals of the learned dynamics from timeseries observations such as video inputs…arXiv cs.LG · 1 h ago · reads at arXiv cs.LG

Exponential Wake-Up Overshoot in Explicit Gaussian Point-Process FiltersarXiv:2610.35958v1 Announce Type: new Abstract: When events resume after a long silence, explicit Gaussian point-process filters amplify the first update exponentially in the silence duration. We give the exact distribution of…arXiv cs.LG · 1 h ago · reads at arXiv cs.LG

This page holds the headline and a short snippet as published by arXiv cs.LG, whose feed they came from. The story itself, the photography and the rights stay with them, and the button above goes straight to their site. We keep this page so the headline has a stable address and a record of when it ran.