Papers › Towards Understanding Why Label Smoothing Degrades Selective Classification and How to Fix It

Towards Understanding Why Label Smoothing Degrades Selective Classification and How to Fix It

19 Mar 2024arXiv:2403.14715archive 2025-07-28

Guoxuan Xia, Olivier Laurent, Gianni Franchi, Christos-Savvas Bouganis

Label smoothing (LS) is a popular regularisation method for training neural networks as it is effective in improving test accuracy and is simple to implement. Hard one-hot labels are smoothed by uniformly distributing probability mass to other classes, reducing overfitting. Prior work has suggested that in some cases LS can degrade selective classification (SC) -- where the aim is to reject misclassifications using a model's uncertainty. In this work, we first demonstrate empirically across an extended range of large-scale tasks and architectures that LS consistently degrades SC. We then address a gap in existing knowledge, providing an explanation for this behaviour by analysing logit-level gradients: LS degrades the uncertainty rank ordering of correct vs incorrect predictions by regularising the max logit more when a prediction is likely to be correct, and less when it is likely to be wrong. This elucidates previously reported experimental results where strong classifiers underperform in SC. We then demonstrate the empirical effectiveness of post-hoc logit normalisation for recovering lost SC performance caused by LS. Furthermore, linking back to our gradient analysis, we again provide an explanation for why such normalisation is effective.

PaperPDFCodeCode 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 10 code samples harvested from 3 repositories linked to this paper: 9 ran (of which 2 constructed an object rather than computing a result; 6 with no instrument failure: 1 honoured, 0 violated, 5 with no contract checked; 3 where Syntology's instrument failed) · 1 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 1 sample from 1 repository, 1 ran (0 with no instrument failure; 1 where Syntology's instrument failed); found in paper text by Syntology: 9 samples from 2 repositories, 8 ran (6 with no instrument failure; 2 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.

No sample is linked to this paper from 3 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

ENSTA-U2IS-AI/Label-smoothing-Selective-classification-Code officialmentioned on GitHubpytorchSyntology: 1 sample, 1 ran (1 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

10 samples harvested: 9 ran (of which 2 constructed an object rather than computing a result; 6 with no instrument failure: 1 honoured, 0 violated, 5 with no contract checked; 3 where Syntology's instrument failed) · 1 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · honoured contract
3ran · our draft was wrong
5ran
1unverified

Licence: 0 of the 10 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 3 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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”. For 1 sample that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “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. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

ENSTA-U2IS-AI/Label-smoothing-Selective-classification-Code official repository 1 sample, 1 ran (1 instrument-failed) · 0 unverified
norm_logits torch_uncertainty_ls/utils.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 2fd3af75ac5f143b · report fixture: signature logits:Tensor/2/float, p:int/0/int, add_unit:float/0/float · returned Tensor · level recorded 2026-09-22
facebookresearch/deit found in paper text by Syntology 2 samples, 2 ran (0 instrument-failed) · 0 unverified
Block in 252 papers
census: 5 implementations from 5 papers, 3 distinct outputs across 2 buckets, one shared input per bucket
models_v2.py ran · metamorphic tier: deterministic Apache-2.0 (permissive) · b02d29318271a1fa · report fixture: returned Block · level recorded 2026-09-22
vit_models models_v2.py ran · metamorphic tier: deterministic Apache-2.0 (permissive) · 067d430bb76eafdd · report fixture: returned vit_models · level recorded 2026-09-22
tensorflow/tpu found in paper text by Syntology 7 samples, 6 ran (2 instrument-failed) · 1 unverified
cross_replica_average models/official/mask_rcnn/tpu_normalization.py ran · our draft was wrong Apache-2.0 (permissive) · 1ab5d66bd8c493da · report fixture: signature t:SymbolicTensor/2/float, num_groups:int/0/int · returned SymbolicTensor · level recorded 2026-09-24
dataset in 10 papers models/official/mnist/mnist_tpu.py ran · our draft was wrong Apache-2.0 (permissive) · 2e4c38986c252157 · report fixture: signature records_file:str/0/str · returned DatasetV1Adapter · level recorded 2026-09-24
efficientnet_x_params models/official/efficientnet/tpu/efficientnet_x_builder.py ran · honoured contract Apache-2.0 (permissive) · 45dc547f9acc69d7 · report fixture: signature model_name:str/0/str · returned tuple · level recorded 2026-09-24
eval_input_fn models/samples/core/get_started/iris_data_tpu.py ran Apache-2.0 (permissive) · 78b2922704a5ab6c · report fixture: signature features:dict/0/?, labels:ndarray/1/int, batch_size:int/0/int · returned DatasetV1Adapter · level recorded 2026-09-24
metric_fn models/official/mnist/mnist_tpu.py ran Apache-2.0 (permissive) · 868a20fd4649650a · report fixture: signature labels:SymbolicTensor/1/int, logits:SymbolicTensor/2/float · returned dict · level recorded 2026-09-24
train_input_fn models/samples/core/get_started/iris_data_tpu.py ran Apache-2.0 (permissive) · 9e48fa6fe2385a5c · report fixture: signature features:dict/0/?, labels:ndarray/1/int, batch_size:int/0/int · returned DatasetV1Adapter · level recorded 2026-09-24
load_data in 700 papers models/samples/core/get_started/iris_data_tpu.py unverified Apache-2.0 (permissive) · 72bd61b1c3a52a02 · 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. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

2 assemblies on this paper: 2 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

2 sibling assemblies for post-hoc logit normalisation (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for post-hoc logit normalisation (graph method: post-hoc logit normalisation). 1 brick; individually on a fixture: 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. norm_logits entry · level 1 · ran · our draft was wrong · 2fd3af75ac5f143b

harvested Assembly for post-hoc logit normalisation (graph method: post-hoc logit normalisation). 3 bricks; individually on a fixture: 3 ran (of which 2 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. vit_models entry · level 1 · ran · 067d430bb76eafdd
  2. Attention helper · level 1 · ran · 389c0e1ba8cccd08
  3. Block helper · level 1 · ran · b02d29318271a1fa

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="2403.14715") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Uncertainty Quantification

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