Scholar Feed

Parameter-efficient fine-tuning (LoRA family)

FedEx-LoRA

FedEx-LoRA: Exact Aggregation for Federated and Efficient Fine-Tuning of Foundation Models

Superseded baseline#53 of 1,113 most-superseded · first seen Oct 12, 2024

Superseded — cited as a baseline and beaten by newer methods

2 papers critique it · 2 beat it on benchmarks

What papers say

Verbatim critique sentences, each from a paper that cites FedEx-LoRA as a baseline.

the method substantially increases the communication cost of fine-tuning since the updated model weights also have to be communicated every round.
— Ravan: Multi-Head Low-Rank Adaptation for Federated Fine-Tuning
“This cause extreme communication overhead, which totally eliminates the lightweight advantage of LoRA”
— Communication-Efficient and Accurate Approach for Aggregation in Federated Low-Rank Adaptation

Beaten on benchmarks

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

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Recent methods in the same sub-problem, not yet superseded in the knowledge base — arXiv benchmark leaders, not vetted production recommendations.