Is AdaMoE superseded?

AdaMoE (Mixture-of-experts routing): superseded — cited as a baseline and beaten by newer methods. 0 paper(s) critique it, 2 beat it on benchmarks — #74 of 1370 most-superseded. Sub-problem: cluster led by LLaMA-MoE. Newer alternatives in the same sub-problem include DOT-MoE, ZEDA, ExpertWeaver, Mixture-of-Experts (MoE) Adaptation, Elastic Mixture-of-Experts (EMoE).

Method Drift›Mixture-of-experts routing

Superseded baseline#74 of 1,370 most-superseded

AdaMoE

AdaMoE: Token-Adaptive Routing with Null Experts for Mixture-of-Experts Language Models

Mixture-of-experts routing · first seen Jun 19, 2024

superseded — cited as a baseline and beaten by newer methods

0 papers critique it · 2 beat it on benchmarks

Beaten on benchmarks

Head-to-head results where a newer method reports beating AdaMoE. Values are copied from the source paper's tables — verify against the cited paper.

ZEDA beats AdaMoE · Avg Acc [Qwen3-30B-A3B]
74.2 vs 54.8
Post-Trained MoE Can Skip Half Experts via Self-Distillation
ZEDA beats AdaMoE · Avg Acc [GLM-4.7-Flash]
71.8 vs 57.1
Post-Trained MoE Can Skip Half Experts via Self-Distillation
EMoE beats AdaMoE · AVG [EMoE vs AdaMoE, k'=1.3]
45.68 vs 41.34
Elastic MoE: Unlocking the Inference-Time Scalability of Mixture-of-Experts
EMoE beats AdaMoE · AVG [EMoE vs AdaMoE, k'=2.2]
48.22 vs 47.18
Elastic MoE: Unlocking the Inference-Time Scalability of Mixture-of-Experts
EMoE beats AdaMoE · AVG [EMoE vs AdaMoE, k'=4.2]
49.00 vs 48.54
Elastic MoE: Unlocking the Inference-Time Scalability of Mixture-of-Experts
EMoE beats AdaMoE · AVG [EMoE vs AdaMoE, k'=6.1]
49.50 vs 48.52
Elastic MoE: Unlocking the Inference-Time Scalability of Mixture-of-Experts

Newer alternatives

Recent methods in the same sub-problem, not yet superseded in the knowledge base.