Papers › In or Out? Fixing ImageNet Out-of-Distribution Detection Evaluation

In or Out? Fixing ImageNet Out-of-Distribution Detection Evaluation

1 Jun 2023arXiv:2306.00826archive 2025-07-28

Julian Bitterwolf, Maximilian Müller, Matthias Hein

Out-of-distribution (OOD) detection is the problem of identifying inputs which are unrelated to the in-distribution task. The OOD detection performance when the in-distribution (ID) is ImageNet-1K is commonly being tested on a small range of test OOD datasets. We find that most of the currently used test OOD datasets, including datasets from the open set recognition (OSR) literature, have severe issues: In some cases more than 50% of the dataset contains objects belonging to one of the ID classes. These erroneous samples heavily distort the evaluation of OOD detectors. As a solution, we introduce with NINCO a novel test OOD dataset, each sample checked to be ID free, which with its fine-grained range of OOD classes allows for a detailed analysis of an OOD detector's strengths and failure modes, particularly when paired with a number of synthetic "OOD unit-tests". We provide detailed evaluations across a large set of architectures and OOD detection methods on NINCO and the unit-tests, revealing new insights about model weaknesses and the effects of pretraining on OOD detection performance. We provide code and data at https://github.com/j-cb/NINCO.

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="2306.00826")

Code

Syntology Ran 1 of 19 code samples harvested from 2 repositories linked to this paper; 18 have no recorded run. Of those that ran: 1 ran · fixture could not drive it.

By repository: official repository: 7 samples from 1 repository, 1 ran; community (archive-listed): 12 samples from 1 repository, 0 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

j-cb/ninco officialmentioned in papermentioned on GitHubpytorchMIT report
robertsi20/Conceptualizing-Uncertainty mentioned on GitHubpytorchMIT 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

19 samples harvested; 1 ran; 0 honoured the contract we drafted; 18 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.

1ran · fixture could not drive it
18unverified

Licence: 0 of the 19 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 2 repositories linked to this paper, official or community; each sample names its own and says which. “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.

all_equal j-cb/NINCO/detection_methods.py official repository ran · fixture could not drive it fingerprinted MIT (permissive) · 9d74dcf8506e1770 · report
auroc_ood j-cb/NINCO/utils.py official repository unverified MIT (permissive) · e4dd80a716476cad · report
evaluate_MSP j-cb/NINCO/detection_methods.py official repository unverified MIT (permissive) · 0ce1ea919855558a · report
kl j-cb/NINCO/detection_methods.py official repository unverified MIT (permissive) · d0fbaed8e9101241 · report
noise_pixel_permutation j-cb/NINCO/noise_functions.py official repository unverified MIT (permissive) · 6249856e2176ac62 · report
noise_rademacher j-cb/NINCO/noise_functions.py official repository unverified MIT (permissive) · ccb68389c662d0c5 · report
noise_uniform j-cb/NINCO/noise_functions.py official repository unverified MIT (permissive) · a60d7c4d12e684cd · report
activation_transform robertsi20/Conceptualizing-Uncertainty/experiment_helpers/experiment_helper_functions.py community (archive-listed) unverified MIT (permissive) · 3d6e8aba845cd198 · report
entropy_uncertainty robertsi20/Conceptualizing-Uncertainty/experiment_helpers/lm_experiment_helper_functions.py community (archive-listed) unverified MIT (permissive) · 9e50d4327bb54248 · report
excerpt_fct robertsi20/Conceptualizing-Uncertainty/experiment_helpers/lm_experiment_helper_functions.py community (archive-listed) unverified MIT (permissive) · 1d5e0e0880eb3ef5 · report
filter_dataset_by_classes robertsi20/Conceptualizing-Uncertainty/experiment_helpers/.ipynb_checkpoints/experiment_helper_functions-checkpoint.py community (archive-listed) unverified MIT (permissive) · 479e0fc1207ac1f5 · report
likelyhood robertsi20/Conceptualizing-Uncertainty/Uncertainty_Functions/unc_functions.py community (archive-listed) unverified MIT (permissive) · f2493ae6234060d7 · report
norm robertsi20/Conceptualizing-Uncertainty/concept_helpers/disc_nmf.py community (archive-listed) unverified MIT (permissive) · 5eadcd84551df892 · report
normalize_matrices robertsi20/Conceptualizing-Uncertainty/concept_helpers/disc_nmf.py community (archive-listed) unverified MIT (permissive) · 832caef4ff71bb80 · report
predict_with_uncertainty_batched robertsi20/Conceptualizing-Uncertainty/experiment_helpers/.ipynb_checkpoints/experiment_helper_functions-checkpoint.py community (archive-listed) unverified MIT (permissive) · 1e531fa3df0ff008 · report
preprocess_images_in_batches robertsi20/Conceptualizing-Uncertainty/experiment_helpers/.ipynb_checkpoints/experiment_helper_functions-checkpoint.py community (archive-listed) unverified MIT (permissive) · 9e910a1703b00aee · report
prob_neg robertsi20/Conceptualizing-Uncertainty/Uncertainty_Functions/unc_functions.py community (archive-listed) unverified MIT (permissive) · e13469045ccef417 · report
prob_pos robertsi20/Conceptualizing-Uncertainty/Uncertainty_Functions/unc_functions.py community (archive-listed) unverified MIT (permissive) · 24a16cd8049665b6 · report
uncertainty_matrices robertsi20/Conceptualizing-Uncertainty/experiment_helpers/lm_experiment_helper_functions.py community (archive-listed) unverified MIT (permissive) · f9ffd38002957cca · report

Tasks

Open Set LearningOut of Distribution (OOD) DetectionOut-of-Distribution Detection

Datasets

Introduced by this paper, per the archive.

ImageNet-1k vs NINCONINCO

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Out-of-Distribution Detection ImageNet-1k vs NINCO ViT-B-384 Mahalanobis (pre-trained on IN-21k) AUROC 95.0 #2 of 5 Archive leaderboard report
Out-of-Distribution Detection ImageNet-1k vs NINCO ViT-B-384 Mahalanobis (pre-trained on IN-21k) FPR@95 27.5 #2 of 5 Archive leaderboard report
Out-of-Distribution Detection ImageNet-1k vs NINCO EffNetv2-M Relative Mahalanobis AUROC 88.9 #4 of 5 Archive leaderboard report
Out-of-Distribution Detection ImageNet-1k vs NINCO EffNetv2-M Relative Mahalanobis FPR@95 54.9 #4 of 5 Archive leaderboard report
Out-of-Distribution Detection ImageNet-1k vs NINCO EffNetb7 Relative Cosine Sim AUROC 87.9 #5 of 5 Archive leaderboard report
Out-of-Distribution Detection ImageNet-1k vs NINCO EffNetb7 Relative Cosine Sim FPR@95 53.7 #5 of 5 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

Test

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