AI Paper Atlas

Research map / Large language models / Instruction and fine-tuning

LIFT+: Lightweight Fine-Tuning for Long-Tail Learning

Jiang-Xin Shi, Tong Wei, Yu-Feng Li

TPAMI 2026 · Large language models › Instruction and fine-tuning · approach: Efficient LLM fine-tuning systems · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add structured lightweight fine-tuning (SLF) (ablation)cosine classifier with LA loss · ImageNet-LTaccuracy: 69.3 -> 74.9Improved
add semantic-aware initialization (SAI) (ablation)LIFT+ with SLF · ImageNet-LTaccuracy: 74.9 -> 76.9Improved
add minimalist data augmentation (MDA) (ablation)LIFT+ with SLF and SAI · ImageNet-LTaccuracy: 76.9 -> 77.0Improved
add test-time ensembling (TTE) (ablation)LIFT+ · ImageNet-LTaccuracy: 77.0 -> 78.3Improved
add structured lightweight fine-tuning (SLF) (ablation)cosine classifier with LA loss · Places-LTaccuracy: 43.9 -> 48.7Improved
add semantic-aware initialization (SAI) (ablation)LIFT+ with SLF · Places-LTaccuracy: 48.7 -> 51.0Improved
add test-time ensembling (TTE) (ablation)LIFT+ · Places-LTaccuracy: 51.5 -> 52.1Improved
fully fine-tune the model instead of only the classifierCLIP with balanced LA loss · ImageNet-LTaccuracy: 71.6% vs. 73.6%Hurt
fully fine-tune the model instead of lightweight fine-tuningCLIP · ImageNet-LTaccuracy: 73.2 -> 77.7Hurt
use semantic-aware initialization instead of random initialization (ablation)LIFT+ · ImageNet-LTaccuracy: 76.5 -> 78.3Improved
use cosine classifier (sigma=25) instead of linear classifier (ablation)LIFT+ · iNaturalist 2018accuracy: 76.7 -> 80.5Improved
use LA loss instead of cross-entropy (ablation)LIFT+ · ImageNet-LTaccuracy: 72.1 -> 78.3Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy