SEUF (Mixture-of-experts routing): superseded — cited as a baseline and beaten by newer methods. 2 paper(s) critique it, 1 beat it on benchmarks — #43 of 1370 most-superseded. Sub-problem: cluster led by DeepSeekMoE. Newer alternatives in the same sub-problem include ProbMoE, Grouter, DECO, SPHERE, AIR-MoE.

Mixture-of-experts routing

SEUF

SEUF: Is Unlearning One Expert Enough for Mixture-of-Experts LLMs?

Superseded baseline#43 of 1,370 most-superseded · first seen Nov 27, 2024

Superseded — cited as a baseline and beaten by newer methods

2 papers critique it · 1 beat it on benchmarks

What papers say

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

Consequently, the only prior work on MoE unlearning, SEUF zhang2024seuf, relies on soft regularization penalties and restricted expert updates, which risks incomplete forgetting.

— GRIP: Algorithm-Agnostic Machine Unlearning for Mixture-of-Experts via Geometric Router Constraints

“restricting unlearning to only one expert, as in SEUF, may limit the capacity for unlearning.”

— Routing-Aware Expert Calibration for Machine Unlearning in Mixture-of-Experts Language Models

Beaten on benchmarks

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

TRACE beats SEUF
0.5541 vs 0.4553
MMLU · [DeepSeek-V2-Lite-Chat]
Routing-Aware Expert Calibration for Machine Unlearning in Mixture-of-Experts Language Models

What to use instead

Recent methods in the same sub-problem, not yet superseded in the knowledge base — arXiv benchmark leaders, not vetted production recommendations.