SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2302.13602 · 2023

The Role of Pre-training Data in Transfer Learning

arXiv · PDF · Open in the Atlas

Code that ran

2 of 3 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 2 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2302.13602.json · badge: /api/ran/2302.13602.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
anonymousmlsubmission/datadistributiontransfer canonical (classifier label)
classifier basis: classifier_2025
2 ran · 1 no recorded run
FunctionStatusWhere it lives
get_loss Ran
ran; our contract draft was wrong, not the code
anonymousmlsubmission/datadistributiontransfer/src/models/simclr/models.py
pointer only (licence: NOASSERTION) · get_code("4c34821ae12949e9")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, anonymousmlsubmission/datadistributiontransfer (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_metric_names Ran
ran; our contract draft was wrong, not the code
anonymousmlsubmission/datadistributiontransfer/src/models/simclr/models.py
pointer only (licence: NOASSERTION) · get_code("345ec4da9f331b85")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, anonymousmlsubmission/datadistributiontransfer (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

build_model Not yet run
unverified (harvested, no recorded run)
anonymousmlsubmission/datadistributiontransfer/src/model.py
pointer only (licence: NOASSERTION) · get_code("4e1f6a9710dc6279")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, anonymousmlsubmission/datadistributiontransfer (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 3 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 3 later) and 0 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
build_model 8 other papers
2 dated earlier, 6 later
2208.05592 (2022-08)
Patching open-vocabulary models by interpolating weights
2212.00638 (2022-12)
Finetune like you pretrain: Improved finetuning of zero-shot vision models
2308.01313 (2023-08)
PerceptionCLIP: Visual Classification by Inferring and Conditioning on Contexts
and 5 more not named here
get_metric_names 3 other papers
1 dated earlier, 2 later
136860514 (2022 (year only))
SLIP: Self-supervision meets Language-Image Pre-training
2304.09172 (2023-04)
Hyperbolic Image-Text Representations
2512.12997 (2025-12)
Calibrating Uncertainty for Zero-Shot Adversarial CLIP
get_loss 1 other paper
1 later
2512.12997 (2025-12)
Calibrating Uncertainty for Zero-Shot Adversarial CLIP

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

The transfer learning paradigm of model pre-training and subsequent fine-tuning produces high-accuracy models. While most studies recommend scaling the pre-training size to benefit most from transfer learning, a question remains: what data and method should be used for pre-training? We investigate the impact of pre-training data distribution on the few-shot and full fine-tuning performance using 3 pre-training methods (supervised, contrastive language-image and image-image), 7 pre-training datasets, and 9 downstream datasets. Through extensive controlled experiments, we find that the choice of the pre-training data source is essential for the few-shot transfer, but its role decreases as more data is made available for fine-tuning. Additionally, we explore the role of data curation and examine the trade-offs between label noise and the size of the pre-training dataset. We find that using 2000X more pre-training data from LAION can match the performance of supervised ImageNet pre-training. Furthermore, we investigate the effect of pre-training methods, comparing language-image contrastive vs. image-image contrastive, and find that the latter leads to better downstream accuracy

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2302.13602")
get_code_for_paper("2302.13602")
have("2302.13602")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2302.13602.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2302.13602.svg)](https://syntology.ai/paper/2302.13602)

Connect an agent — have() is free.