LFT (Parameter-efficient fine-tuning (LoRA family)): superseded — cited as a baseline and beaten by newer methods. 1 paper(s) critique it, 1 beat it on benchmarks — #174 of 1113 most-superseded. Sub-problem: cluster led by CNN. Newer alternatives in the same sub-problem include Structured Convolutional Projection + LoRA, LEGO, layer-selective multimodal large language models (MLLMs) with contrastive LoRA tuning and layer sensitivity analysis (LSA), AdaptPrompt.

Method Drift›Parameter-efficient fine-tuning (LoRA family)

Superseded baseline#174 of 1,113 most-superseded

LFT

Parameter-efficient fine-tuning (LoRA family)

superseded — cited as a baseline and beaten by newer methods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 LFT as a baseline.

“However, this assumption may not hold when the target is a geometrically distinct shape family that differs fundamentally from the training distribution, since the frozen layers may not have learned sufficiently general features for the new geometry.”
— Adapting Automotive Aerodynamics Surrogates to New Vehicle Families via Transfer Learning

Beaten on benchmarks

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

LoRA beats LFT · Force Shear R² [leave-one-out]
0.946 vs 0.523
Adapting Automotive Aerodynamics Surrogates to New Vehicle Families via Transfer Learning
LoRA beats LFT · Surface Pressure rel L2 [leave-one-out]
3.21 vs 7.85
Adapting Automotive Aerodynamics Surrogates to New Vehicle Families via Transfer Learning
LoRA beats LFT · Surface Friction rel L2 [leave-one-out]
0.198 vs 0.573
Adapting Automotive Aerodynamics Surrogates to New Vehicle Families via Transfer Learning

Newer alternatives

Recent methods in the same sub-problem, not yet superseded in the knowledge base.