Papers › Long-Tail Learning with Foundation Model: Heavy Fine-Tuning Hurts

Long-Tail Learning with Foundation Model: Heavy Fine-Tuning Hurts

18 Sep 2023arXiv:2309.10019archive 2025-07-28

Jiang-Xin Shi, Tong Wei, Zhi Zhou, Jie-Jing Shao, Xin-Yan Han, Yu-Feng Li

The fine-tuning paradigm in addressing long-tail learning tasks has sparked significant interest since the emergence of foundation models. Nonetheless, how fine-tuning impacts performance in long-tail learning was not explicitly quantified. In this paper, we disclose that heavy fine-tuning may even lead to non-negligible performance deterioration on tail classes, and lightweight fine-tuning is more effective. The reason is attributed to inconsistent class conditions caused by heavy fine-tuning. With the observation above, we develop a low-complexity and accurate long-tail learning algorithms LIFT with the goal of facilitating fast prediction and compact models by adaptive lightweight fine-tuning. Experiments clearly verify that both the training time and the learned parameters are significantly reduced with more accurate predictive performance compared with state-of-the-art approaches. The implementation code is available at https://github.com/shijxcs/LIFT.

PaperPDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2309.10019")

Code

Syntology Ran 5 of 8 code samples harvested from 1 repository linked to this paper; 3 have no recorded run. Of those that ran: 4 ran · our draft was wrong; 1 ran with no contract checked.

By repository: official repository: 8 samples from 1 repository, 5 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

shijxcs/lift officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

8 samples harvested; 5 ran; 0 honoured the contract we drafted; 3 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

4ran · our draft was wrong
1ran
3unverified

Licence: 8 of the 8 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from shijxcs/lift. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

LoRA shijxcs/lift/models/peft_modules.py official repository ran · metamorphic tier: invariant no licence file found · pointer only · 152823b85e68d486 · report
basic_clean shijxcs/LIFT/clip/simple_tokenizer.py official repository ran · our draft was wrong fingerprinted no licence file found · pointer only · 98f385d847636a3e · report
focal_loss shijxcs/LIFT/utils/losses.py official repository ran · our draft was wrong fingerprinted no licence file found · pointer only · 4dcf06ba28983266 · report
get_pairs shijxcs/LIFT/clip/simple_tokenizer.py official repository ran · our draft was wrong no licence file found · pointer only · d919ae32e5e4e616 · report
whitespace_clean shijxcs/LIFT/clip/simple_tokenizer.py official repository ran · our draft was wrong fingerprinted no licence file found · pointer only · 9542161e9640b858 · report
build_model shijxcs/LIFT/clip/model.py official repository unverified no licence file found · pointer only · c47aa9e5b049a11d · report
load shijxcs/LIFT/clip/clip.py official repository unverified no licence file found · pointer only · c22a78c25372b59f · report
load_vit_to_cpu shijxcs/LIFT/trainer.py official repository unverified no licence file found · pointer only · 3597147d80c14dc2 · report

Tasks

Fine-Grained Image ClassificationImage ClassificationLong-tail LearningLong-tail learning with class descriptors

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Long-tail Learning CIFAR-100-LT (ρ=10) LIFT (ViT-B/16, ImageNet-21K pre-training) Error Rate 8.7 #1 of 31 Archive leaderboard report
Long-tail Learning CIFAR-100-LT (ρ=10) LIFT (ViT-B/16, CLIP) Error Rate 15.1 #4 of 31 Archive leaderboard report
Long-tail Learning CIFAR-100-LT (ρ=100) LIFT (ViT-B/16, ImageNet-21K pre-training) Error Rate 10.9 #2 of 66 Archive leaderboard report
Long-tail Learning CIFAR-100-LT (ρ=100) LIFT (ViT-B/16, CLIP) Error Rate 18.3 #3 of 66 Archive leaderboard report
Long-tail Learning CIFAR-100-LT (ρ=50) LIFT (ViT-B/16, ImageNet-21K pre-training) Error Rate 9.8 #1 of 25 Archive leaderboard report
Long-tail Learning CIFAR-100-LT (ρ=50) LIFT (ViT-B/16, CLIP) Error Rate 16.9 #4 of 25 Archive leaderboard report
Long-tail Learning ImageNet-LT LIFT (ViT-L/14) Top-1 Accuracy 82.9 #1 of 69 Archive leaderboard report
Long-tail Learning ImageNet-LT LIFT (ViT-B/16) Top-1 Accuracy 78.3 #4 of 69 Archive leaderboard report
Long-tail Learning Places-LT LIFT (ViT-L/14) Top-1 Accuracy 53.7 #1 of 29 Archive leaderboard report
Long-tail Learning Places-LT LIFT (ViT-B/16) Top-1 Accuracy 52.2 #2 of 29 Archive leaderboard report
Long-tail Learning iNaturalist 2018 LIFT (ViT-L/14@336px) Top-1 Accuracy 87.4% #1 of 43 Archive leaderboard report
Long-tail Learning iNaturalist 2018 LIFT (ViT-L/14) Top-1 Accuracy 85.2% #2 of 43 Archive leaderboard report
Long-tail Learning iNaturalist 2018 LIFT (ViT-B/16) Top-1 Accuracy 80.4% #5 of 43 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

CLIP

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections