Papers › Do ImageNet-trained models learn shortcuts? The impact of frequency shortcuts on generalization

Do ImageNet-trained models learn shortcuts? The impact of frequency shortcuts on generalization

5 Mar 2025CVPR 2025 1arXiv:2503.03519archive 2025-07-28

Shunxin Wang, Raymond Veldhuis, Nicola Strisciuglio

Frequency shortcuts refer to specific frequency patterns that models heavily rely on for correct classification. Previous studies have shown that models trained on small image datasets often exploit such shortcuts, potentially impairing their generalization performance. However, existing methods for identifying frequency shortcuts require expensive computations and become impractical for analyzing models trained on large datasets. In this work, we propose the first approach to more efficiently analyze frequency shortcuts at a larger scale. We show that both CNN and transformer models learn frequency shortcuts on ImageNet. We also expose that frequency shortcut solutions can yield good performance on out-of-distribution (OOD) test sets which largely retain texture information. However, these shortcuts, mostly aligned with texture patterns, hinder model generalization on rendition-based OOD test sets. These observations suggest that current OOD evaluations often overlook the impact of frequency shortcuts on model generalization. Future benchmarks could thus benefit from explicitly assessing and accounting for these shortcuts to build models that generalize across a broader range of OOD scenarios.

PaperPDFConference PDFCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 18 code samples harvested from 1 repository linked to this paper: 14 ran (of which 1 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 10 where Syntology's instrument failed) · 4 unverified. Where Syntology's instrument failed: 9 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: found in paper text by Syntology: 18 samples from 1 repository, 14 ran (4 with no instrument failure; 10 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-22 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

nis-research/hfss found in paper text by SyntologySyntology: 18 samples, 14 ran (10 instrument-failed) · 4 unverified 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

18 samples harvested: 14 ran (of which 1 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 10 where Syntology's instrument failed) · 4 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · honoured contract
9ran · our draft was wrong
1ran; the contract check could not run
3ran
4unverified

Licence: 0 of the 18 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 nis-research/hfss. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (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 are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. 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 six 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; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. 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.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

nis-research/hfss found in paper text by Syntology 18 samples, 14 ran (10 instrument-failed) · 4 unverified
White_Mask hfss/LSI.py ran MIT (permissive) · 93097b2e633389ff · report fixture: returned White_Mask · level recorded 2026-09-22
dict_to_namedtuple hfss/LSI.py ran · our draft was wrong MIT (permissive) · 7dade86e6e8016ce · report fixture: signature d:dict/0/? · returned Args · level recorded 2026-09-22
filter in 20 papers hfss/LSI.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 3609e233a3276554 · report fixture: signature x:Tensor/3/complex, mask:Tensor/2/complex · returned Tensor · level recorded 2026-09-22
gen_freqs_list hfss/LSI.py ran · honoured contract fingerprinted MIT (permissive) · 052937d76a92f757 · report fixture: signature h:int/0/int, w:int/0/int · returned list · level recorded 2026-09-22
generate_mask in 24 papers hfss/LSI.py ran · our draft was wrong MIT (permissive) · 2e0170b8a1504ca2 · report fixture: signature sel_freqs:list/1/tuple, h:int/0/int, w:int/0/int · returned Tensor · level recorded 2026-09-22
get_candidates in 12 papers hfss/LSI.py ran · our draft was wrong MIT (permissive) · f23b96953d0f71ed · report fixture: signature args:Args/0/?, num_candidates:int/0/int, portion:float/0/float, stage:str/0/str · returned list · level recorded 2026-09-22
get_dataloader in 111 papers hfss/LSI.py ran MIT (permissive) · 302eb65e3d7d2b08 · report fixture: signature args:ArgsConfig/0/?, dataset:SimpleDataset/0/?, shuffle:bool/0/bool, pin_memory:bool/0/bool · returned DataLoader · level recorded 2026-09-22
get_next_subpolicy hfss/LSI.py ran · our draft was wrong MIT (permissive) · 605bc1e05087f4bb · report fixture: signature args:Args/0/?, transform_candidates:list/1/callable, b:int/0/int · returned tuple · level recorded 2026-09-22
get_topn_subpolicies_classwise hfss/LSI.py ran · our draft was wrong MIT (permissive) · 2d37e6f004585e77 · report fixture: signature args:Args/0/?, subpolicies:list/1/tuple, N:int/0/int · returned list · level recorded 2026-09-22
parse_args in 355 papers hfss/LSI.py ran · our draft was wrong MIT (permissive) · ea5692df62e9082f · report fixture: signature kwargs:dict/0/? · returned tuple · level recorded 2026-09-22
sample_frequency hfss/LSI.py ran · our draft was wrong fingerprinted MIT (permissive) · 1eecfa5954a49671 · report fixture: signature portion:float/0/float, f_list:list/1/float · returned list · level recorded 2026-09-22
split_dataset in 65 papers hfss/LSI.py ran · our draft was wrong MIT (permissive) · 5f913213824f631e · report fixture: signature args:ArgsConfig/0/?, dataset:SimpleDataset/0/?, k:int/0/int, chosen_class:NoneType/0/? · returned tuple · level recorded 2026-09-22
validate_child hfss/LSI.py ran MIT (permissive) · ad6933c7dd4d745f · report fixture: returned tuple · level recorded 2026-09-22
validate_classwise hfss/LSI.py ran · our draft was wrong MIT (permissive) · 55e40597f228b3fc · report fixture: signature args:Args/0/?, model:callable/0/?, criterion:callable/0/?, valid_loader:list/1/tuple, device:NoneType/0/? · returned tuple · level recorded 2026-09-22
LFSI hfss/LSI.py unverified MIT (permissive) · 25d9b9da400444d3 · report
get_dataset in 285 papers hfss/LSI.py unverified MIT (permissive) · afd66aeb8288064b · report
process_fn in 5 papers hfss/LSI.py unverified MIT (permissive) · 2223e054c99bf8d0 · report
search_subpolicies hfss/LSI.py unverified MIT (permissive) · 36874b7691bd1679 · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for approach to more efficiently analyze frequency shortcuts (graph method: approach to more efficiently analyze frequency shortcuts). 18 bricks; individually on a fixture: 14 ran (of which 1 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 10 where Syntology's instrument failed) · 4 unverified. The assembly has not been run end-to-end.

  1. LFSI entry · level 0 · unverified · 25d9b9da400444d3
  2. get_candidates helper · level 1 · ran · our draft was wrong · f23b96953d0f71ed
  3. get_dataset helper · level 0 · unverified · afd66aeb8288064b
  4. process_fn helper · level 0 · unverified · 2223e054c99bf8d0
  5. split_dataset helper · level 1 · ran · our draft was wrong · 5f913213824f631e
  6. White_Mask helper · level 1 · ran · 93097b2e633389ff
  7. gen_freqs_list helper · level 1 · ran · honoured contract · 052937d76a92f757
  8. generate_mask data_transform · level 1 · ran · our draft was wrong · 2e0170b8a1504ca2
  9. sample_frequency sampler · level 1 · ran · our draft was wrong · 1eecfa5954a49671
  10. get_topn_subpolicies_classwise helper · level 1 · ran · our draft was wrong · 2d37e6f004585e77
  11. parse_args helper · level 1 · ran · our draft was wrong · ea5692df62e9082f
  12. search_subpolicies helper · level 0 · unverified · 36874b7691bd1679
  13. filter data_transform · level 1 · ran; the contract check could not run (RAISES) · 3609e233a3276554
  14. dict_to_namedtuple helper · level 1 · ran · our draft was wrong · 7dade86e6e8016ce
  15. get_next_subpolicy helper · level 1 · ran · our draft was wrong · 605bc1e05087f4bb
  16. validate_child metric · level 1 · ran · ad6933c7dd4d745f
  17. get_dataloader helper · level 1 · ran · 302eb65e3d7d2b08
  18. validate_classwise helper · level 1 · ran · our draft was wrong · 55e40597f228b3fc

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2503.03519") returns these assemblies with each brick's role and wiring (how to connect).

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

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