Papers › Exploring Target Representations for Masked Autoencoders

Exploring Target Representations for Masked Autoencoders

8 Sep 2022arXiv:2209.03917archive 2025-07-28

Xingbin Liu, Jinghao Zhou, Tao Kong, Xianming Lin, Rongrong Ji

Masked autoencoders have become popular training paradigms for self-supervised visual representation learning. These models randomly mask a portion of the input and reconstruct the masked portion according to the target representations. In this paper, we first show that a careful choice of the target representation is unnecessary for learning good representations, since different targets tend to derive similarly behaved models. Driven by this observation, we propose a multi-stage masked distillation pipeline and use a randomly initialized model as the teacher, enabling us to effectively train high-capacity models without any efforts to carefully design target representations. Interestingly, we further explore using teachers of larger capacity, obtaining distilled students with remarkable transferring ability. On different tasks of classification, transfer learning, object detection, and semantic segmentation, the proposed method to perform masked knowledge distillation with bootstrapped teachers (dBOT) outperforms previous self-supervised methods by nontrivial margins. We hope our findings, as well as the proposed method, could motivate people to rethink the roles of target representations in pre-training masked autoencoders.The code and pre-trained models are publicly available at https://github.com/liuxingbin/dbot.

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="2209.03917")

Code

Syntology Ran 5 of 13 code samples harvested from 1 repository linked to this paper; 8 have no recorded run. Of those that ran: 1 ran · honoured contract; 1 ran · fixture could not drive it; 3 ran with no contract checked.

By repository: official repository: 13 samples from 1 repository, 5 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

liuxingbin/dbot officialmentioned in papermentioned on GitHubpytorchApache-2.0 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

13 samples harvested; 5 ran; 1 honoured the contract we drafted; 8 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 · honoured contract
1ran · fixture could not drive it
3ran
8unverified

Licence: 0 of the 13 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 liuxingbin/dbot. “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.

bool_flag liuxingbin/dbot/evaluation/utils.py official repository ran Apache-2.0 (permissive) · c19231378e41dc5a · report
clip_gradients liuxingbin/dbot/evaluation/utils.py official repository ran Apache-2.0 (permissive) · 2b5489053b222e00 · report
cosine_scheduler liuxingbin/dbot/evaluation/utils.py official repository ran · honoured contract fingerprinted Apache-2.0 (permissive) · 361a6b24f11fc50a · report
drop_path liuxingbin/dbot/evaluation/models.py official repository ran · fixture could not drive it fingerprinted Apache-2.0 (permissive) · 55120f2026b56aa2 · report
get_num_layer_for_vit liuxingbin/dbot/evaluation/optim_factory.py official repository ran fingerprinted Apache-2.0 (permissive) · a3448702d6025ce6 · report
build_dataset liuxingbin/dbot/evaluation/eval_cls.py official repository unverified Apache-2.0 (permissive) · ac882d65ee949545 · report
build_transform liuxingbin/dbot/evaluation/eval_cls.py official repository unverified Apache-2.0 (permissive) · 1902796bef81ffd2 · report
compute_distance_matrix liuxingbin/dbot/analysis/analysis.py official repository unverified Apache-2.0 (permissive) · 336aa445b9756b05 · report
compute_mean_attention_dist liuxingbin/dbot/analysis/analysis.py official repository unverified Apache-2.0 (permissive) · b77522e9d6eaa3af · report
get_dataset liuxingbin/dbot/analysis/analysis.py official repository unverified Apache-2.0 (permissive) · 481536e2fedc4873 · report
get_parameter_groups liuxingbin/dbot/evaluation/optim_factory.py official repository unverified Apache-2.0 (permissive) · bb32ca015a494c58 · report
vit_small liuxingbin/dbot/analysis/vision_transformer.py official repository unverified Apache-2.0 (permissive) · 48bb4f68134b5f20 · report
vit_tiny liuxingbin/dbot/analysis/vision_transformer.py official repository unverified Apache-2.0 (permissive) · 6769eebdd8702e50 · report

Tasks

Image ClassificationInstance SegmentationKnowledge DistillationObject DetectionRepresentation LearningSelf-Supervised Image ClassificationSemantic SegmentationTransfer Learningobject-detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Classification ImageNet dBOT ViT-H (CLIP as Teacher) Top 1 Accuracy 88.2% #57 of 1060 Archive leaderboard report
Image Classification ImageNet dBOT ViT-L (CLIP as Teacher) Top 1 Accuracy 87.8% #68 of 1060 Archive leaderboard report
Image Classification ImageNet dBOT ViT-B (CLIP as Teacher) Top 1 Accuracy 85.7% #208 of 1060 Archive leaderboard report
Instance Segmentation COCO test-dev dBOT ViT-L (CLIP) mask AP 48.8 #26 of 112 Archive leaderboard report
Instance Segmentation COCO test-dev dBOT ViT-L mask AP 48.3 #29 of 112 Archive leaderboard report
Instance Segmentation COCO test-dev dBOT ViT-B mask AP 46.3 #36 of 112 Archive leaderboard report
Instance Segmentation COCO test-dev dBOT ViT-B (CLIP) mask AP 46.2 #37 of 112 Archive leaderboard report
Object Detection COCO test-dev dBOT ViT-L (CLIP) box mAP 56.8 #39 of 225 Archive leaderboard report
Object Detection COCO test-dev dBOT ViT-L box mAP 56.1 #45 of 225 Archive leaderboard report
Object Detection COCO test-dev dBOT ViT-B (CLIP) box mAP 53.6 #59 of 225 Archive leaderboard report
Object Detection COCO test-dev dBOT ViT-B box mAP 53.5 #62 of 225 Archive leaderboard report
Self-Supervised Image Classification ImageNet (finetuned) dBOT (ViT-H/14) Number of Params 632M #6 of 65 Archive leaderboard report
Self-Supervised Image Classification ImageNet (finetuned) dBOT (ViT-H/14) Top 1 Accuracy 88.0% #6 of 65 Archive leaderboard report
Semantic Segmentation ADE20K dBOT ViT-L (CLIP) Validation mIoU 56.2 #42 of 235 Archive leaderboard report
Semantic Segmentation ADE20K dBOT ViT-L Validation mIoU 55.2 #49 of 235 Archive leaderboard report
Semantic Segmentation ADE20K dBOT ViT-B (CLIP) Validation mIoU 52.9 #85 of 235 Archive leaderboard report
Semantic Segmentation ADE20K dBOT ViT-B Validation mIoU 50.8 #109 of 235 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

Knowledge Distillation

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