Papers › LDReg: Local Dimensionality Regularized Self-Supervised Learning

LDReg: Local Dimensionality Regularized Self-Supervised Learning

19 Jan 2024arXiv:2401.10474archive 2025-07-28

Hanxun Huang, Ricardo J. G. B. Campello, Sarah Monazam Erfani, Xingjun Ma, Michael E. Houle, James Bailey

Representations learned via self-supervised learning (SSL) can be susceptible to dimensional collapse, where the learned representation subspace is of extremely low dimensionality and thus fails to represent the full data distribution and modalities. Dimensional collapse also known as the "underfilling" phenomenon is one of the major causes of degraded performance on downstream tasks. Previous work has investigated the dimensional collapse problem of SSL at a global level. In this paper, we demonstrate that representations can span over high dimensional space globally, but collapse locally. To address this, we propose a method called local dimensionality regularization (LDReg). Our formulation is based on the derivation of the Fisher-Rao metric to compare and optimize local distance distributions at an asymptotically small radius for each data point. By increasing the local intrinsic dimensionality, we demonstrate through a range of experiments that LDReg improves the representation quality of SSL. The results also show that LDReg can regularize dimensionality at both local and global levels.

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 36 code samples harvested from 2 repositories linked to this paper: 20 ran (of which 5 constructed an object rather than computing a result; 12 with no instrument failure: 4 honoured, 1 violated, 7 with no contract checked; 8 where Syntology's instrument failed) · 16 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; 6 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 6 samples from 1 repository, 5 ran (1 with no instrument failure; 4 where Syntology's instrument failed); community (archive-listed): 30 samples from 1 repository, 15 ran (11 with no instrument failure; 4 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.

HanxunH/LDReg officialmentioned in papermentioned on GitHubpytorchSyntology: 6 samples, 5 ran (4 instrument-failed) · 1 unverified report
JoJoNing25/DDAug mentioned on GitHubpytorchSyntology: 30 samples, 15 ran (4 instrument-failed) · 15 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

36 samples harvested: 20 ran (of which 5 constructed an object rather than computing a result; 12 with no instrument failure: 4 honoured, 1 violated, 7 with no contract checked; 8 where Syntology's instrument failed) · 16 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.

4ran · honoured contract
1ran · violated contract
2ran · our draft was wrong
6ran; the contract check could not run
7ran
16unverified

Licence: 0 of the 36 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; 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 2 samples 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.

HanxunH/LDReg official repository 6 samples, 5 ran (4 instrument-failed) · 1 unverified
NTXentBase losses/ntxent_lid_reg.py ran MIT (permissive) · f3d6a6be48cdb9e9 · report fixture: returned NTXentBase · level recorded 2026-09-22
accuracy in 872 papers
census: 8 implementations from 7 papers, 7 distinct outputs across 6 buckets, one shared input per bucket
losses/ntxent_lid_reg.py ran; the contract check could not run (RAISES) MIT (permissive) · e8d9c82a15ba2a57 · report fixture: signature output:Tensor/2/float, target:Tensor/1/int, topk:tuple/1/int · returned list · level recorded 2026-09-22
gather in 75 papers losses/ntxent_lid_reg.py ran · our draft was wrong MIT (permissive) · cf0782bfe6eddbac · report fixture: signature input:Tensor/2/float · returned tuple · level recorded 2026-09-22
lid_mle losses/ntxent_lid_reg.py ran; the contract check could not run (RAISES) MIT (permissive) · 9d8f2e35d4702814 · report fixture: signature data:Tensor/2/float, reference:Tensor/2/float, k:int/0/int, get_idx:bool/0/bool, compute_mode:str/0/str · returned Tensor · level recorded 2026-09-22
lid_mom_est losses/ntxent_lid_reg.py ran; the contract check could not run (RAISES) MIT (permissive) · 918cecffdec2a619 · report fixture: signature data:Tensor/2/float, reference:Tensor/2/float, k:int/0/int, get_idx:bool/0/bool, compute_mode:str/0/str · returned Tensor · level recorded 2026-09-22
NTXentLIDReg losses/ntxent_lid_reg.py unverified MIT (permissive) · 0cfb402f67dc5765 · report
JoJoNing25/DDAug community (archive-listed) 30 samples, 15 ran (4 instrument-failed) · 15 unverified
FullGatherLayer losses/ntxent_lidaug.py ran MIT (permissive) · efbec016cf3c4e21 · report fixture: returned FullGatherLayer · level recorded 2026-09-22
GrayLayer losses/ntxent_lidaug.py ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · d03096e93a9cffcc · report fixture: returned GrayLayer · level recorded 2026-09-22
IdentityLayer in 7 papers losses/ntxent_lidaug.py ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · dcb240bf1416737e · report fixture: returned IdentityLayer · level recorded 2026-09-22
_STE losses/ntxent_lidaug.py ran MIT (permissive) · 0948c0465513c79d · report fixture: returned _STE · level recorded 2026-09-22
_blend_image losses/ntxent_lidaug.py ran · honoured contract fingerprinted MIT (permissive) · 3bccd0b613b2ec94 · report fixture: signature img1:Tensor/4/float, img2:Tensor/4/float, alpha:Tensor/1/float · returned Tensor · level recorded 2026-09-22
_blur losses/ntxent_lidaug.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted MIT (permissive) · 00b297b23155ccd0 · report fixture: signature img:Tensor/4/float, kernel:Tensor/2/float · returned Tensor · level recorded 2026-09-22
_gray losses/ntxent_lidaug.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted MIT (permissive) · 14a3e0d0584f4f1f · report fixture: signature img:Tensor/4/float · returned Tensor · level recorded 2026-09-22
activation in 30 papers losses/ntxent_lidaug.py ran · honoured contract fingerprinted MIT (permissive) · cd6af889cd96c427 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
eye_rank losses/ntxent_lidaug.py ran · violated contract fingerprinted MIT (permissive) · 4593546c432138ea · report fixture: signature n:int/0/int, device:NoneType/0/? · returned Tensor · level recorded 2026-09-22
get_sharpness_kernel losses/ntxent_lidaug.py ran · honoured contract MIT (permissive) · a77358b38f7eb68c · report fixture: signature device:device/0/? · returned Tensor · level recorded 2026-09-22
lid_mle losses/ntxent_lidaug.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted MIT (permissive) · 397aafa38e8e9bfa · report fixture: signature data:Tensor/2/float, reference:Tensor/2/float, k:int/0/int, get_idx:bool/0/bool · returned Tensor · level recorded 2026-09-22
rank in 18 papers
census: 3 implementations from 5 papers, 3 distinct outputs across 2 buckets, one shared input per bucket
losses/ntxent_lidaug.py ran · metamorphic tier: well formed MIT (permissive) · a38b33a0219b2651 · report fixture: returned int · level recorded 2026-09-22
ste losses/ntxent_lidaug.py ran · honoured contract fingerprinted MIT (permissive) · f8ccb18e75fab7db · report fixture: signature input_forward:Tensor/2/float, input_backward:Tensor/3/float · returned Tensor · level recorded 2026-09-22
tensor_function losses/ntxent_lidaug.py ran · our draft was wrong MIT (permissive) · 229e3f0e1a7ceb04 · report fixture: signature func:callable/0/? · returned function · level recorded 2026-09-22
world_size losses/ntxent_lidaug.py ran · metamorphic tier: well formed MIT (permissive) · a1d11ea20c97a5b5 · report fixture: returned int · level recorded 2026-09-22
AugLayer losses/ntxent_lidaug.py unverified MIT (permissive) · 8758e55ff0b748ef · report
AugPolicy losses/ntxent_lidaug.py unverified MIT (permissive) · 99feec893219b44d · report
BrightnessLayer losses/ntxent_lidaug.py unverified MIT (permissive) · 5872bbeca7913d9c · report
ContrastLayer losses/ntxent_lidaug.py unverified MIT (permissive) · c566f6618abd39ed · report
GaussianBlurLayer losses/ntxent_lidaug.py unverified MIT (permissive) · 83b86abab1e05654 · report
HueLayer losses/ntxent_lidaug.py unverified MIT (permissive) · 057cf4bcc8260a07 · report
NTXentLossLIDAug losses/ntxent_lidaug.py unverified MIT (permissive) · 5887614dd9999cbd · report
PosterizeLayer losses/ntxent_lidaug.py unverified MIT (permissive) · 712473be9274437d · report
SaturationLayer losses/ntxent_lidaug.py unverified MIT (permissive) · c2f2a5f8fcf3adac · report
SharpnessLayer losses/ntxent_lidaug.py unverified MIT (permissive) · 7b025f0f036c7b4c · report
SolarizeLayer losses/ntxent_lidaug.py unverified MIT (permissive) · ef94ac6f4485c581 · report
full_gather losses/ntxent_lidaug.py unverified MIT (permissive) · 18ac34711a665335 · report
posterize in 7 papers losses/ntxent_lidaug.py unverified MIT (permissive) · 01c6078e675ae833 · report
sharpness in 9 papers losses/ntxent_lidaug.py unverified MIT (permissive) · 959973c5950dd3fb · report
solarize in 15 papers losses/ntxent_lidaug.py unverified MIT (permissive) · 8000f04f9e183df4 · 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 LDReg (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for LDReg (graph method: LDReg). 30 bricks; individually on a fixture: 15 ran (of which 4 constructed an object rather than computing a result; 11 with no instrument failure: 4 honoured, 1 violated, 6 with no contract checked; 4 where Syntology's instrument failed) · 15 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. NTXentLossLIDAug entry · level 0 · unverified · 5887614dd9999cbd
  2. AugPolicy helper · level 0 · unverified · 99feec893219b44d
  3. eye_rank helper · level 1 · ran · violated contract · 4593546c432138ea
  4. full_gather helper · level 0 · unverified · 18ac34711a665335
  5. lid_mle helper · level 1 · ran; the contract check could not run (RAISES) · 397aafa38e8e9bfa
  6. rank helper · level 1 · ran · a38b33a0219b2651
  7. world_size helper · level 1 · ran · a1d11ea20c97a5b5
  8. AugLayer helper · level 0 · unverified · 8758e55ff0b748ef
  9. FullGatherLayer helper · level 1 · ran · efbec016cf3c4e21
  10. BrightnessLayer helper · level 0 · unverified · 5872bbeca7913d9c
  11. ContrastLayer helper · level 0 · unverified · c566f6618abd39ed
  12. GaussianBlurLayer helper · level 0 · unverified · 83b86abab1e05654
  13. GrayLayer helper · level 1 · ran · d03096e93a9cffcc
  14. HueLayer helper · level 0 · unverified · 057cf4bcc8260a07
  15. IdentityLayer helper · level 1 · ran · dcb240bf1416737e
  16. PosterizeLayer helper · level 0 · unverified · 712473be9274437d
  17. SaturationLayer helper · level 0 · unverified · c2f2a5f8fcf3adac
  18. SharpnessLayer helper · level 0 · unverified · 7b025f0f036c7b4c
  19. SolarizeLayer helper · level 0 · unverified · ef94ac6f4485c581
  20. _STE helper · level 1 · ran · 0948c0465513c79d
  21. _blend_image helper · level 1 · ran · honoured contract · 3bccd0b613b2ec94
  22. _blur helper · level 1 · ran; the contract check could not run (RAISES) · 00b297b23155ccd0
  23. _gray helper · level 1 · ran; the contract check could not run (RAISES) · 14a3e0d0584f4f1f
  24. activation helper · level 1 · ran · honoured contract · cd6af889cd96c427
  25. get_sharpness_kernel helper · level 1 · ran · honoured contract · a77358b38f7eb68c
  26. ste helper · level 1 · ran · honoured contract · f8ccb18e75fab7db
  27. tensor_function helper · level 1 · ran · our draft was wrong · 229e3f0e1a7ceb04
  28. posterize helper · level 0 · unverified · 01c6078e675ae833
  29. sharpness helper · level 0 · unverified · 959973c5950dd3fb
  30. solarize helper · level 0 · unverified · 8000f04f9e183df4

harvested Assembly for LDReg (graph method: LDReg). 10 bricks; individually on a fixture: 9 ran (of which 2 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 1 violated, 4 with no contract checked; 4 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. NTXentLIDReg entry · level 0 · unverified · 0cfb402f67dc5765
  2. NTXentBase loss · level 1 · ran · f3d6a6be48cdb9e9
  3. accuracy helper · level 1 · ran; the contract check could not run (RAISES) · e8d9c82a15ba2a57
  4. gather helper · level 1 · ran · our draft was wrong · cf0782bfe6eddbac
  5. lid_mle helper · level 1 · ran; the contract check could not run (RAISES) · 9d8f2e35d4702814
  6. lid_mom_est helper · level 1 · ran; the contract check could not run (RAISES) · 918cecffdec2a619
  7. eye_rank helper · level 1 · ran · violated contract · 4593546c432138ea
  8. rank helper · level 1 · ran · a38b33a0219b2651
  9. world_size helper · level 1 · ran · a1d11ea20c97a5b5
  10. FullGatherLayer helper · level 1 · ran · efbec016cf3c4e21

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="2401.10474") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Self-Supervised Learning

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