BTM (Mixture-of-experts routing): superseded — cited as a baseline and beaten by newer methods. 1 paper(s) critique it, 1 beat it on benchmarks — #99 of 1370 most-superseded. Sub-problem: cluster led by BTX. Newer alternatives in the same sub-problem include MetaMoE, BAR, BERT-MoE Framework, null experts within token-choice MoE, MixtureKit.

Superseded baseline#99 of 1,370 most-superseded

BTM

Mixture-of-experts routing

superseded — cited as a baseline and beaten by newer methods

1 papers critique it · 1 beat it on benchmarks

What papers say

Verbatim critique sentences, each from a paper that cites BTM as a baseline.

“BTM ensembling demonstrates strong performance; however, the inference cost increases linearly with the number of experts relative to a single dense model.”
— Training-Free Dynamic Upcycling of Expert Language Models

Beaten on benchmarks

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

\methodName beats BTM · Average Accuracy [CLIP ViT-B/32]
94.52 vs 90.33
MetaMoE: Diversity-Aware Proxy Selection for Privacy-Preserving Mixture-of-Experts Unification
\methodName beats BTM · Average Accuracy [CLIP ViT-B/16]
96.24 vs 91.75
MetaMoE: Diversity-Aware Proxy Selection for Privacy-Preserving Mixture-of-Experts Unification
\methodName beats BTM · Average Accuracy [LLaMA-3.2-3B]
74.42 vs 72.74
MetaMoE: Diversity-Aware Proxy Selection for Privacy-Preserving Mixture-of-Experts Unification
\methodName beats BTM · Average Accuracy [LLaMA-3.1-8B]
81.59 vs 79.79
MetaMoE: Diversity-Aware Proxy Selection for Privacy-Preserving Mixture-of-Experts Unification

Newer alternatives

Recent methods in the same sub-problem, not yet superseded in the knowledge base.