Papers › Towards Latent Masked Image Modeling for Self-Supervised Visual Representation Learning

Towards Latent Masked Image Modeling for Self-Supervised Visual Representation Learning

22 Jul 2024arXiv:2407.15837archive 2025-07-28

Yibing Wei, Abhinav Gupta, Pedro Morgado

Masked Image Modeling (MIM) has emerged as a promising method for deriving visual representations from unlabeled image data by predicting missing pixels from masked portions of images. It excels in region-aware learning and provides strong initializations for various tasks, but struggles to capture high-level semantics without further supervised fine-tuning, likely due to the low-level nature of its pixel reconstruction objective. A promising yet unrealized framework is learning representations through masked reconstruction in latent space, combining the locality of MIM with the high-level targets. However, this approach poses significant training challenges as the reconstruction targets are learned in conjunction with the model, potentially leading to trivial or suboptimal solutions.Our study is among the first to thoroughly analyze and address the challenges of such framework, which we refer to as Latent MIM. Through a series of carefully designed experiments and extensive analysis, we identify the source of these challenges, including representation collapsing for joint online/target optimization, learning objectives, the high region correlation in latent space and decoding conditioning. By sequentially addressing these issues, we demonstrate that Latent MIM can indeed learn high-level representations while retaining the benefits of MIM models.

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 29 code samples harvested from 1 repository linked to this paper: 21 ran (of which 10 constructed an object rather than computing a result; 18 with no instrument failure: 5 honoured, 0 violated, 13 with no contract checked; 3 where Syntology's instrument failed) · 8 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: 29 samples from 1 repository, 21 ran (18 with no instrument failure; 3 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.

yibingwei-1/latentmim officialmentioned in papermentioned on GitHubpytorchSyntology: 29 samples, 21 ran (3 instrument-failed) · 8 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

29 samples harvested: 21 ran (of which 10 constructed an object rather than computing a result; 18 with no instrument failure: 5 honoured, 0 violated, 13 with no contract checked; 3 where Syntology's instrument failed) · 8 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.

5ran · honoured contract
3ran · our draft was wrong
13ran
8unverified

Licence: 29 of the 29 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 yibingwei-1/latentmim. “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 6 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.

yibingwei-1/latentmim official repository 29 samples, 21 ran (3 instrument-failed) · 8 unverified
AbsolutePositionEmbeds models_lmim.py ran no licence file found · pointer only · 14d22cad6c3652df · report fixture: returned AbsolutePositionEmbeds · level recorded 2026-09-22
Attention in 344 papers
census: 13 implementations from 12 papers, 10 distinct outputs across 4 buckets, one shared input per bucket
models_lmim.py ran no licence file found · pointer only · 97fb93bd98cfce58 · report fixture: returned Attention · level recorded 2026-09-22
Block in 252 papers
census: 5 implementations from 5 papers, 3 distinct outputs across 2 buckets, one shared input per bucket
models_lmim.py ran no licence file found · pointer only · 88d3a859bbb353fa · report fixture: returned Block · level recorded 2026-09-22
CrossAttention in 85 papers models_lmim.py ran no licence file found · pointer only · 0aa2506ff4669ad9 · report fixture: returned CrossAttention · level recorded 2026-09-22
CrossBlock models_lmim.py ran no licence file found · pointer only · 09c1e9e18a95d73c · report fixture: returned CrossBlock · level recorded 2026-09-22
CrossPositionAttentionStem models_lmim.py ran no licence file found · pointer only · 94c80da22c9b99f5 · report fixture: returned CrossPositionAttentionStem · level recorded 2026-09-22
PatchEmbed in 123 papers
census: 9 implementations from 9 papers, 6 distinct outputs on one shared input
models_lmim.py ran no licence file found · pointer only · 8936931f961fac53 · report fixture: returned PatchEmbed · level recorded 2026-09-22
Patchify models_lmim.py ran no licence file found · pointer only · e6455c9712d66617 · report fixture: returned Patchify · level recorded 2026-09-22
Similarity in 5 papers models_lmim.py ran fingerprinted no licence file found · pointer only · 630c0648178f5810 · report fixture: returned Similarity · level recorded 2026-09-22
SinCosPEs models_lmim.py ran fingerprinted no licence file found · pointer only · e287eb398f89f6ed · report fixture: returned SinCosPEs · level recorded 2026-09-22
_build_mlp models_lmim.py ran · our draft was wrong no licence file found · pointer only · d4031567204db418 · report fixture: signature in_dim:int/0/int, out_dim:int/0/int, hidden_dim:int/0/int, num_layers:int/0/int, norm:callable/0/?, out_norm:bool/0/bool · returned Sequential · level recorded 2026-09-22
adjust_target_momentum main_lmim.py ran · honoured contract no licence file found · pointer only · 4a3f7a34b495299c · report fixture: signature epoch:int/0/int, args:Args/0/? · returned float · level recorded 2026-09-24
all_reduce_mean in 27 papers util/misc.py ran · honoured contract fingerprinted no licence file found · pointer only · 3dc19396537db789 · report fixture: signature x:float/0/float · returned float · level recorded 2026-09-24
avg_pairwise_sim models_lmim.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 1fe90f92040cf053 · report fixture: signature q:Tensor/3/float, k:Tensor/3/float · returned Tensor · level recorded 2026-09-22
concat_all_gather in 141 papers
census: 7 implementations from 7 papers, 2 distinct outputs on one shared input
util/dist_utils.py ran · honoured contract fingerprinted no licence file found · pointer only · 84c0965ec39be82d · report fixture: signature tensor:Tensor/2/float · returned Tensor · level recorded 2026-09-24
concat_all_gather in 141 papers
census: 7 implementations from 7 papers, 2 distinct outputs on one shared input
util/misc.py ran · honoured contract fingerprinted no licence file found · pointer only · 8cacb1e15bf1d910 · report fixture: signature tensor:Tensor/2/float · returned Tensor · level recorded 2026-09-24
cosine_decay_schedule main_lmim.py ran fingerprinted no licence file found · pointer only · f25dd0ec378284fd · report fixture: signature ep:int/0/int, epochs:int/0/int, init_val:float/0/float, end_val:float/0/float · returned float · level recorded 2026-09-24
generate_fewshot_dataset data_tools.py ran no licence file found · pointer only · 44c6311e21193939 · report fixture: signature data:list/1/Datum, num_shots:int/0/int, repeat:bool/0/bool, seed:int/0/int · returned list · level recorded 2026-09-24
get_grad_norm_ in 22 papers util/misc.py ran · honoured contract no licence file found · pointer only · ba1356e8ceb654d2 · report fixture: signature parameters:list/1/Tensor, norm_type:float/0/float · returned Tensor · level recorded 2026-09-25
read_and_split_data data_tools.py ran no licence file found · pointer only · aa97db87276a3a8e · report fixture: signature image_dir:str/0/str, p_trn:float/0/float, p_val:float/0/float, ignored:tuple/1/empty, new_cnames:NoneType/0/? · returned tuple · level recorded 2026-09-24
read_splits data_tools.py ran · our draft was wrong no licence file found · pointer only · f71fd7e968578cce · report fixture: signature filepath:str/0/str, path_prefix:str/0/str · returned tuple · level recorded 2026-09-24
CrossDecoder models_lmim.py unverified no licence file found · pointer only · f9294385468d37e7 · report
LMIM models_lmim.py unverified no licence file found · pointer only · e102fd2444a9a5ae · report
VIT models_lmim.py unverified no licence file found · pointer only · 9f056dcd045ee6cf · report
adjust_learning_rate in 125 papers util/lr_sched.py unverified no licence file found · pointer only · 80500372237a4794 · report
build_vit vits.py unverified no licence file found · pointer only · b80212c1206d00ab · report
imagenet datasets.py unverified no licence file found · pointer only · 975cbd8043721e2c · report
imagenet100 datasets.py unverified no licence file found · pointer only · 2479657c36a797a0 · report
load_dataset in 276 papers datasets.py unverified no licence file found · pointer only · 980a8e933845de91 · 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 Latent Masked Image Modeling (graph method: Latent Masked Image Modeling). 15 bricks; individually on a fixture: 12 ran (of which 10 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 2 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

  1. LMIM entry · level 0 · unverified · e102fd2444a9a5ae
  2. CrossDecoder model · level 0 · unverified · f9294385468d37e7
  3. PatchEmbed model · level 1 · ran · 8936931f961fac53
  4. Patchify model · level 1 · ran · e6455c9712d66617
  5. VIT model · level 0 · unverified · 9f056dcd045ee6cf
  6. _build_mlp helper · level 1 · ran · our draft was wrong · d4031567204db418
  7. avg_pairwise_sim metric · level 1 · ran · our draft was wrong · 1fe90f92040cf053
  8. AbsolutePositionEmbeds helper · level 1 · ran · 14d22cad6c3652df
  9. CrossBlock model · level 1 · ran · 09c1e9e18a95d73c
  10. CrossPositionAttentionStem model · level 1 · ran · 94c80da22c9b99f5
  11. Block model · level 1 · ran · 88d3a859bbb353fa
  12. SinCosPEs helper · level 1 · ran · e287eb398f89f6ed
  13. Attention model · level 1 · ran · 97fb93bd98cfce58
  14. CrossAttention helper · level 1 · ran · 0aa2506ff4669ad9
  15. Similarity helper · level 1 · ran · 630c0648178f5810

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="2407.15837") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Representation Learning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

MIM

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