Signals 4 · free daily AI digest

How to Loop MoE: Flatten the Experts, Untie the Attention

Paper recorded by Signals 4 on 2026-09-28 in cs.AI. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-28 on arXiv · recorded by Signals 4 on 2026-09-29

Category: cs.AI · 人工智能 · first seen 2026-09-29

Abstract

Looped Transformers reuse one block of layers several times: by spending extra computation they push a model of fixed size further, and so use its parameters more fully; while sparse mixture-of-experts (MoE) models activate only a few of many experts for each token. Looped MoE bridges these two design philosophies and gives MoE models new potential for better expert usage, but it raises a question

Read on arXiv →

#5 most recent of 440 cs.AI papers we have recorded · ↑ newer: TokenCast: Forecasting Token Consumption During LLM Agent Execution · ↓ older: KV-streams for Efficient Compaction in Agentic Reinforcement Learning
Cite this page: How to Loop MoE: Flatten the Experts, Untie the Attention: the #5 most recent of 440 cs.AI papers we have recorded (as of 2026-09-28). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/how-to-loop-moe-flatten-the-experts-untie-the-attention.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable: papers.json
Related: More cs.AI papers · arXiv signals · All papers · Today in AI
Get 4 AI signals a day by email — free.
Subscribe free → See all plans →
Get 4 AI signals a day by email — free
All models · All repos · By company · Daily editions