arXiv AI September 3, 2026 By Lucas Qingyang Fang, Tiyao Liu, Jinhao Jing, Zeji Li, Kaijie Chen, Harikrishna Kuttivelil, Katia Obraczka

Convergence Theory of Knowledge Distillation in Asynchronous P2P Gossip Learning Network

Read the original on arXiv AI →

arXiv:2609. 01952v1 Announce Type: cross Abstract: Decentralized, serverless learning increasingly connects devices running different architectures, where the standard tool, decentralized SGD, is undefined as models with different parameter counts cannot be averaged.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv Machine Learning

Oct 21

Discount Yourself: Self-inclusive Freshness-weighted Gossip for Decentralised Learning under Client Outages

arXiv:2610.28545v1 Announce Type: new Abstract: Decentralised federated learning (DFL) trains personalised models on devices that average with their neighbours rather than with a central server. Real...

By Qiyuan Wang, Ke Xiao, Christos Anagnostopoulos

benchmarks

Hugging Face Trending Papers

Oct 20

Discount Yourself: Self-inclusive Freshness-weighted Gossip for Decentralised Learning under Client Outages

Decentralised federated learning (DFL) trains personalised models on devices that average with their neighbours rather than with a central server. Real devices, however, are often offline for hours or...

benchmarks

arXiv Machine Learning

Aug 10

Stream Learning: Partition-Fair Gossip Learning Without Tokens

arXiv:2608. 06946v1 Announce Type: cross Abstract: In gossip learning, a network of nodes trains a shared model collaboratively, without a central coordinator, by repeatedly exchanging parts of their local models.

By Fabien Mathieu (NPA), Alexandre Pham (NPA), Maria Gradinariu Potop-Butucaru (NPA), S{\'e}bastien Tixeuil (IUF, NPA)

benchmarks safety

arXiv Machine Learning

Aug 18

The Distributional View of Knowledge Distillation

arXiv:2608. 15215v1 Announce Type: cross Abstract: Token-level knowledge distillation (KD) matches two conditional distributions per position, yet the standard objectives compare them pointwise: a Kullback-Leibler gradient is blind to which wrong token receives probability mass.

By Gordei Verbii, Juho Lee

rag fine-tuning efficiency

arXiv AI

Sep 1

Knowledge Distillation under Teacher Misspecification: An Order-Parameter Analysis of the Gap between Teacher Mimicry and Task Performance

arXiv:2608.29472v1 Announce Type: cross Abstract: Knowledge distillation trains a small student model to reproduce the outputs of a large teacher model, and its progress is typically monitored throug...

By Kazuyuki Hara, Hideitsu Hino

efficiency

arXiv Machine Learning

Oct 7

Uniform Discrete Diffusion Models are Minimax Optimal for Estimating Distributions with Small Effective Support Size

arXiv:2610.07655v1 Announce Type: cross Abstract: Discrete diffusion models have emerged as a practically successful framework for generative modeling on discrete product spaces, yet their statistica...

By Dongsun Yoon, Saptarshi Chakraborty

diffusion

One email a morning, machine-written

Related stories

Discount Yourself: Self-inclusive Freshness-weighted Gossip for Decentralised Learning under Client Outages

Discount Yourself: Self-inclusive Freshness-weighted Gossip for Decentralised Learning under Client Outages

Stream Learning: Partition-Fair Gossip Learning Without Tokens

The Distributional View of Knowledge Distillation

Knowledge Distillation under Teacher Misspecification: An Order-Parameter Analysis of the Gap between Teacher Mimicry and Task Performance

Uniform Discrete Diffusion Models are Minimax Optimal for Estimating Distributions with Small Effective Support Size