Papers › The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization

The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization

29 Jun 2020ICCV 2021 10arXiv:2006.16241archive 2025-07-28

Dan Hendrycks, Steven Basart, Norman Mu, Saurav Kadavath, Frank Wang, Evan Dorundo, Rahul Desai, Tyler Zhu, Samyak Parajuli, Mike Guo, Dawn Song, Jacob Steinhardt, Justin Gilmer

We introduce four new real-world distribution shift datasets consisting of changes in image style, image blurriness, geographic location, camera operation, and more. With our new datasets, we take stock of previously proposed methods for improving out-of-distribution robustness and put them to the test. We find that using larger models and artificial data augmentations can improve robustness on real-world distribution shifts, contrary to claims in prior work. We find improvements in artificial robustness benchmarks can transfer to real-world distribution shifts, contrary to claims in prior work. Motivated by our observation that data augmentations can help with real-world distribution shifts, we also introduce a new data augmentation method which advances the state-of-the-art and outperforms models pretrained with 1000 times more labeled data. Overall we find that some methods consistently help with distribution shifts in texture and local image statistics, but these methods do not help with some other distribution shifts like geographic changes. Our results show that future research must study multiple distribution shifts simultaneously, as we demonstrate that no evaluated method consistently improves robustness.

PaperPDFConference PDFCodeCode 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 14 code samples harvested from 1 repository linked to this paper: 14 ran (of which 1 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 12 where Syntology's instrument failed) · 0 unverified. Where Syntology's instrument failed: 12 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 14 samples from 1 repository, 14 ran (2 with no instrument failure; 12 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 between 2026-09-22 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

hendrycks/imagenet-r officialmentioned in papermentioned on GitHubpytorchSyntology: 14 samples, 14 ran (12 instrument-failed) · 0 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

14 samples harvested: 14 ran (of which 1 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 12 where Syntology's instrument failed) · 0 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.

12ran · our draft was wrong
2ran

Licence: 0 of the 14 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 hendrycks/imagenet-r. “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.

hendrycks/imagenet-r official repository 14 samples, 14 ran (12 instrument-failed) · 0 unverified
AugMix deepfashion_remix/augmix.py ran MIT (permissive) · 90f0b6606b7bec55 · report fixture: returned AugMix · level recorded 2026-09-22
autocontrast in 9 papers deepfashion_remix/augmix.py ran · our draft was wrong MIT (permissive) · 48502588e268dd8f · report fixture: signature pil_img:Image/0/?, level:NoneType/0/? · returned Image · level recorded 2026-09-22
brightness deepfashion_remix/augmix.py ran · our draft was wrong MIT (permissive) · 7fa83d532dab3fa4 · report fixture: signature pil_img:Image/0/?, level:int/0/int · returned Image · level recorded 2026-09-22
color in 9 papers deepfashion_remix/augmix.py ran · our draft was wrong MIT (permissive) · f53b32f9516fdaa1 · report fixture: signature pil_img:Image/0/?, level:int/0/int · returned Image · level recorded 2026-09-22
contrast deepfashion_remix/augmix.py ran · our draft was wrong MIT (permissive) · 6e50ec13656740c3 · report fixture: signature pil_img:Image/0/?, level:int/0/int · returned Image · level recorded 2026-09-22
equalize in 10 papers deepfashion_remix/augmix.py ran · our draft was wrong MIT (permissive) · cbb8d1e8526fa7eb · report fixture: signature pil_img:Image/0/?, level:NoneType/0/? · returned Image · level recorded 2026-09-22
posterize in 7 papers deepfashion_remix/augmix.py ran · our draft was wrong MIT (permissive) · e51e7aefc03823aa · report fixture: signature pil_img:Image/0/?, level:int/0/int · returned Image · level recorded 2026-09-22
rotate in 56 papers deepfashion_remix/augmix.py ran · our draft was wrong MIT (permissive) · f3de03c4a896f7e5 · report fixture: signature pil_img:Image/0/?, level:int/0/int · returned Image · level recorded 2026-09-22
sharpness in 9 papers deepfashion_remix/augmix.py ran · our draft was wrong MIT (permissive) · c55e7427481ee18e · report fixture: signature pil_img:Image/0/?, level:int/0/int · returned Image · level recorded 2026-09-22
shear_x in 43 papers deepfashion_remix/augmix.py ran · our draft was wrong MIT (permissive) · 8b301c76d5b96b5f · report fixture: signature pil_img:Image/0/?, level:int/0/int · returned Image · level recorded 2026-09-22
shear_y in 37 papers deepfashion_remix/augmix.py ran · our draft was wrong MIT (permissive) · 74de255c18b450ef · report fixture: signature pil_img:Image/0/?, level:int/0/int · returned Image · level recorded 2026-09-22
solarize in 15 papers deepfashion_remix/augmix.py ran · our draft was wrong MIT (permissive) · ea3db097ab0c59d3 · report fixture: signature pil_img:Image/0/?, level:int/0/int · returned Image · level recorded 2026-09-22
translate_x in 7 papers deepfashion_remix/augmix.py ran · our draft was wrong MIT (permissive) · 68ee08ee193d291c · report fixture: signature pil_img:Image/0/?, level:int/0/int · returned Image · level recorded 2026-09-22
translate_y in 7 papers deepfashion_remix/augmix.py ran MIT (permissive) · 49792e8ad13a4ec5 · report fixture: signature pil_img:Image/0/?, level:int/0/int · returned Image · level recorded 2026-09-25

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 new data augmentation method (graph method: new data augmentation method). 18 bricks; individually on a fixture: 18 ran (of which 1 constructed an object rather than computing a result; 6 with no instrument failure: 3 honoured, 1 violated, 2 with no contract checked; 12 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

  1. AugMix entry · level 1 · ran · 90f0b6606b7bec55
  2. autocontrast data_transform · level 1 · ran · our draft was wrong · 48502588e268dd8f
  3. brightness data_transform · level 1 · ran · our draft was wrong · 7fa83d532dab3fa4
  4. color data_transform · level 1 · ran · our draft was wrong · f53b32f9516fdaa1
  5. contrast data_transform · level 1 · ran · our draft was wrong · 6e50ec13656740c3
  6. equalize data_transform · level 1 · ran · our draft was wrong · cbb8d1e8526fa7eb
  7. posterize data_transform · level 1 · ran · our draft was wrong · e51e7aefc03823aa
  8. rotate data_transform · level 1 · ran · our draft was wrong · f3de03c4a896f7e5
  9. sharpness data_transform · level 1 · ran · our draft was wrong · c55e7427481ee18e
  10. shear_x data_transform · level 1 · ran · our draft was wrong · 8b301c76d5b96b5f
  11. shear_y data_transform · level 1 · ran · our draft was wrong · 74de255c18b450ef
  12. solarize data_transform · level 1 · ran · our draft was wrong · ea3db097ab0c59d3
  13. translate_x data_transform · level 1 · ran · our draft was wrong · 68ee08ee193d291c
  14. translate_y data_transform · level 1 · ran · 49792e8ad13a4ec5
  15. float_parameter helper · level 1 · ran · honoured contract · d6ad456eb790edde
  16. sample_level helper · level 1 · ran · violated contract · 5cc8d4764ac07a35
  17. int_parameter helper · level 1 · ran · honoured contract · cce32b359dda4b20
  18. rand_lvl helper · level 1 · ran · honoured contract · 4eaeb136b1e28c9e

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="2006.16241") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Data AugmentationDomain GeneralizationOut-of-Distribution Generalization

Datasets

Introduced by this paper, per the archive.

ImageNet-R

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Domain Generalization ImageNet-C DeepAugment (ResNet-50) mean Corruption Error (mCE) 60.4 #36 of 47 Archive leaderboard report
Domain Generalization ImageNet-R DeepAugment+AugMix (ResNet-50) Top-1 Error Rate 53.2 #29 of 39 Archive leaderboard report
Domain Generalization ImageNet-R DeepAugment (ResNet-50) Top-1 Error Rate 57.8 #33 of 39 Archive leaderboard report
Domain Generalization VizWiz-Classification ResNet-50 (deepaugment) Accuracy - All Images 41.3 #30 of 90 Archive leaderboard report
Domain Generalization VizWiz-Classification ResNet-50 (deepaugment) Accuracy - Clean Images 46 #30 of 90 Archive leaderboard report
Domain Generalization VizWiz-Classification ResNet-50 (deepaugment) Accuracy - Corrupted Images 34.9 #30 of 90 Archive leaderboard report
Domain Generalization VizWiz-Classification ResNet-50 (deepaugment+augmix) Accuracy - All Images 40.3 #35 of 90 Archive leaderboard report
Domain Generalization VizWiz-Classification ResNet-50 (deepaugment+augmix) Accuracy - Clean Images 44.5 #35 of 90 Archive leaderboard report
Domain Generalization VizWiz-Classification ResNet-50 (deepaugment+augmix) Accuracy - Corrupted Images 34.1 #35 of 90 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.

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