Papers › ESC: Erasing Space Concept for Knowledge Deletion

ESC: Erasing Space Concept for Knowledge Deletion

3 Apr 2025CVPR 2025 1arXiv:2504.02199archive 2025-07-28

Tae-Young Lee, Sundong Park, Minwoo Jeon, Hyoseok Hwang, Gyeong-Moon Park

As concerns regarding privacy in deep learning continue to grow, individuals are increasingly apprehensive about the potential exploitation of their personal knowledge in trained models. Despite several research efforts to address this, they often fail to consider the real-world demand from users for complete knowledge erasure. Furthermore, our investigation reveals that existing methods have a risk of leaking personal knowledge through embedding features. To address these issues, we introduce a novel concept of Knowledge Deletion (KD), an advanced task that considers both concerns, and provides an appropriate metric, named Knowledge Retention score (KR), for assessing knowledge retention in feature space. To achieve this, we propose a novel training-free erasing approach named Erasing Space Concept (ESC), which restricts the important subspace for the forgetting knowledge by eliminating the relevant activations in the feature. In addition, we suggest ESC with Training (ESC-T), which uses a learnable mask to better balance the trade-off between forgetting and preserving knowledge in KD. Our extensive experiments on various datasets and models demonstrate that our proposed methods achieve the fastest and state-of-the-art performance. Notably, our methods are applicable to diverse forgetting scenarios, such as facial domain setting, demonstrating the generalizability of our methods. The code is available at http://github.com/KU-VGI/ESC .

PaperPDFConference PDFCodeCode 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 24 code samples harvested from 1 repository linked to this paper: 10 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 1 honoured, 0 violated, 5 with no contract checked; 4 where Syntology's instrument failed) · 14 unverified. Where Syntology's instrument failed: 4 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 24 samples from 1 repository, 10 ran (6 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.

ku-vgi/esc officialmentioned in papermentioned on GitHubpytorchSyntology: 24 samples, 10 ran (4 instrument-failed) · 14 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

24 samples harvested: 10 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 1 honoured, 0 violated, 5 with no contract checked; 4 where Syntology's instrument failed) · 14 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.

1ran · honoured contract
4ran · our draft was wrong
5ran
14unverified

Licence: 24 of the 24 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 ku-vgi/esc. “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”. “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.

ku-vgi/esc official repository 24 samples, 10 ran (4 instrument-failed) · 14 unverified
AllCNN unlearn.py ran · metamorphic tier: invariant no licence file found · pointer only · e8d79e942930f478 · report fixture: signature x:Tensor/4/float, all:bool/0/bool · returned Tensor · level recorded 2026-09-22
ComplexDNN unlearn.py ran · metamorphic tier: invariant no licence file found · pointer only · e141c147ec7315da · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
Conv in 30 papers
census: 8 implementations from 8 papers, 8 distinct outputs across 2 buckets, one shared input per bucket this sample: bucket 1 of 2, arg 1: rank 4, kind float, dtype float32
unlearn.py ran · metamorphic tier: invariant fingerprinted no licence file found · pointer only · d574992e1f3b5967 · report fixture: signature input:Tensor/4/float · returned Tensor · level recorded 2026-09-22
CustomDataset unlearn.py ran · metamorphic tier: deterministic fingerprinted no licence file found · pointer only · 2dd17365b6d40bcc · report fixture: signature index:int/0/int · returned tuple · level recorded 2026-09-22
Flatten in 23 papers
census: 9 implementations from 9 papers, 2 distinct outputs across 2 buckets, one shared input per bucket this sample: bucket 1 of 2, arg 1: rank 4, kind float, dtype float32
unlearn.py ran · metamorphic tier: invariant fingerprinted no licence file found · pointer only · 0b787ec506d382c7 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
compute_losses in 9 papers unlearn.py ran · honoured contract no licence file found · pointer only · 232f710ba2a3df96 · report fixture: signature net:callable/0/?, loader:list/1/tuple, device:str/0/str · returned ndarray · level recorded 2026-09-22
get_unlearn_loader unlearn.py ran · our draft was wrong no licence file found · pointer only · 08afcdeb55afcff8 · report fixture: signature trainset:SimpleDataset/0/?, testset:SimpleDataset/0/?, test_trainset:SimpleDataset/0/?, forget_class:list/1/int, batch_size:int/0/int, num_forget:int/0/int, repair_num_ratio:float/0/float · returned tuple · level recorded 2026-09-22
load_vit unlearn.py ran · our draft was wrong no licence file found · pointer only · 6dc9eb0c21a32237 · report fixture: signature model_name:str/0/str, num_classes:int/0/int, device:str/0/str, is_pretrained:bool/0/bool, is_backbone_freezed:bool/0/bool · returned VisionTransformer · level recorded 2026-09-22
optimizer_picker unlearn.py ran · our draft was wrong no licence file found · pointer only · 0c6538eed7fe129a · report fixture: signature optimization:str/0/str, param:generator/0/?, lr:float/0/float, momentum:float/0/float · returned Adam · level recorded 2026-09-22
split_class_data unlearn.py ran · our draft was wrong no licence file found · pointer only · bf1ccc66583510a7 · report fixture: signature dataset:SimpleDataset/0/?, forget_class:list/1/int, num_forget:int/0/int · returned tuple · level recorded 2026-09-22
all_eval unlearn.py unverified no licence file found · pointer only · 27672850dd40958b · report
create_dir in 10 papers unlearn.py unverified no licence file found · pointer only · 54f51ff8da6ff35e · report
eval in 90 papers unlearn.py unverified no licence file found · pointer only · c6e8179c8f9ea281 · report
evaluate_KR unlearn.py unverified no licence file found · pointer only · 5ccd63f514797a72 · report
evaluate_mia unlearn.py unverified no licence file found · pointer only · 03f1d47518a9d3e3 · report
evaluate_model in 54 papers unlearn.py unverified no licence file found · pointer only · 8083bb6afa3f1008 · report
exp_summary unlearn.py unverified no licence file found · pointer only · cc2022b7a141973c · report
get_dataset in 285 papers unlearn.py unverified no licence file found · pointer only · 99c85c380ca6c5cf · report
main in 235 papers unlearn.py unverified no licence file found · pointer only · a31522fb137cd85a · report
prepare_dataset in 32 papers unlearn.py unverified no licence file found · pointer only · e98c82942ca364b2 · report
seed_torch unlearn.py unverified no licence file found · pointer only · 0e55b39d4f65c5a8 · report
simple_mia unlearn.py unverified no licence file found · pointer only · 1aadf59595786019 · report
simple_mia_pytorch unlearn.py unverified no licence file found · pointer only · 1aa81e0d0aebdfa1 · report
train_model in 103 papers unlearn.py unverified no licence file found · pointer only · ad05cd5a51f7a47c · 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 Erasing Space Concept (ESC) (graph method: Erasing Space Concept (ESC)). 25 bricks; individually on a fixture: 11 ran (of which 1 constructed an object rather than computing a result; 7 with no instrument failure: 1 honoured, 0 violated, 6 with no contract checked; 4 where Syntology's instrument failed) · 14 unverified. The assembly has not been run end-to-end.

  1. main entry · level 0 · unverified · a31522fb137cd85a
  2. AllCNN helper · level 1 · ran · e8d79e942930f478
  3. all_eval helper · level 0 · unverified · 27672850dd40958b
  4. create_dir helper · level 0 · unverified · 54f51ff8da6ff35e
  5. evaluate_KR helper · level 0 · unverified · 5ccd63f514797a72
  6. evaluate_mia helper · level 0 · unverified · 03f1d47518a9d3e3
  7. exp_summary helper · level 0 · unverified · cc2022b7a141973c
  8. load_vit helper · level 1 · ran · our draft was wrong · 6dc9eb0c21a32237
  9. prepare_dataset data_transform · level 0 · unverified · e98c82942ca364b2
  10. seed_torch helper · level 0 · unverified · 0e55b39d4f65c5a8
  11. Conv helper · level 1 · ran · d574992e1f3b5967
  12. Flatten helper · level 1 · ran · 0b787ec506d382c7
  13. Identity helper · level 1 · ran · cd43c253a0e41d2f
  14. eval helper · level 0 · unverified · c6e8179c8f9ea281
  15. optimizer_picker helper · level 1 · ran · our draft was wrong · 0c6538eed7fe129a
  16. compute_losses helper · level 1 · ran · honoured contract · 232f710ba2a3df96
  17. simple_mia helper · level 0 · unverified · 1aadf59595786019
  18. simple_mia_pytorch helper · level 0 · unverified · 1aa81e0d0aebdfa1
  19. get_dataset helper · level 0 · unverified · 99c85c380ca6c5cf
  20. get_unlearn_loader helper · level 1 · ran · our draft was wrong · 08afcdeb55afcff8
  21. ComplexDNN helper · level 1 · ran · e141c147ec7315da
  22. evaluate_model helper · level 0 · unverified · 8083bb6afa3f1008
  23. train_model helper · level 0 · unverified · ad05cd5a51f7a47c
  24. CustomDataset helper · level 1 · ran · 2dd17365b6d40bcc
  25. split_class_data helper · level 1 · ran · our draft was wrong · bf1ccc66583510a7

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="2504.02199") returns these assemblies with each brick's role and wiring (how to connect).

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