Papers › WISE: Rethinking the Knowledge Memory for Lifelong Model Editing of Large Language Models

WISE: Rethinking the Knowledge Memory for Lifelong Model Editing of Large Language Models

23 May 2024arXiv:2405.14768archive 2025-07-28

Peng Wang, Zexi Li, Ningyu Zhang, Ziwen Xu, Yunzhi Yao, Yong Jiang, Pengjun Xie, Fei Huang, Huajun Chen

Large language models (LLMs) need knowledge updates to meet the ever-growing world facts and correct the hallucinated responses, facilitating the methods of lifelong model editing. Where the updated knowledge resides in memories is a fundamental question for model editing. In this paper, we find that editing either long-term memory (direct model parameters) or working memory (non-parametric knowledge of neural network activations/representations by retrieval) will result in an impossible triangle -- reliability, generalization, and locality can not be realized together in the lifelong editing settings. For long-term memory, directly editing the parameters will cause conflicts with irrelevant pretrained knowledge or previous edits (poor reliability and locality). For working memory, retrieval-based activations can hardly make the model understand the edits and generalize (poor generalization). Therefore, we propose WISE to bridge the gap between memories. In WISE, we design a dual parametric memory scheme, which consists of the main memory for the pretrained knowledge and a side memory for the edited knowledge. We only edit the knowledge in the side memory and train a router to decide which memory to go through when given a query. For continual editing, we devise a knowledge-sharding mechanism where different sets of edits reside in distinct subspaces of parameters, and are subsequently merged into a shared memory without conflicts. Extensive experiments show that WISE can outperform previous model editing methods and overcome the impossible triangle under lifelong model editing of question answering, hallucination, and out-of-distribution settings across trending LLM architectures, e.g., GPT, LLaMA, and Mistral. Code is available at https://github.com/zjunlp/EasyEdit.

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 16 code samples harvested from 1 repository linked to this paper: 13 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 9 where Syntology's instrument failed) · 3 unverified. Where Syntology's instrument failed: 3 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: 16 samples from 1 repository, 13 ran (4 with no instrument failure; 9 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 1 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.

zjunlp/easyedit officialmentioned in papermentioned on GitHubpytorchMITSyntology: 16 samples, 13 ran (9 instrument-failed) · 3 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

16 samples harvested: 13 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 9 where Syntology's instrument failed) · 3 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
3ran · our draft was wrong
6ran; the contract check could not run
3ran
3unverified

Licence: 0 of the 16 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 zjunlp/easyedit. “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.

zjunlp/easyedit official repository 16 samples, 13 ran (9 instrument-failed) · 3 unverified
EditingMeanAct easyeditor/models/wise/WISE.py ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · 6556782be86e66b2 · report fixture: signature val:float/0/float · returned NoneType · level recorded 2026-09-22
bernoulli in 8 papers easyeditor/models/wise/WISE.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 7e7272686fd1e3c2 · report fixture: signature tensor:Tensor/2/float, density:float/0/float, rescale:bool/0/bool · returned Tensor · level recorded 2026-09-22
euc easyeditor/models/wise/WISE.py ran · our draft was wrong MIT (permissive) · e6c1e18615ecb8af · report fixture: signature query:Tensor/3/float, key:Tensor/3/float, config:MinimalConfig/0/?, act_mask:Tensor/2/float, infer:bool/0/bool · returned Tensor · level recorded 2026-09-24
get_mask in 119 papers
census: 9 implementations from 10 papers, 9 distinct outputs across 7 buckets, one shared input per bucket this sample: bucket 2 of 7, arg 1: rank 2, kind float, dtype float32
easyeditor/models/wise/WISE.py ran · our draft was wrong MIT (permissive) · 5bb8c583c306754b · report fixture: signature delta:Tensor/2/float, method:str/0/str, mask_dtype:dtype/0/? · returned Tensor · level recorded 2026-09-22
get_task_vectors easyeditor/models/wise/WISE.py ran · our draft was wrong MIT (permissive) · a0efe0cf5f74ca6c · report fixture: signature base:Tensor/2/float, tensors:list/1/Tensor · returned tuple · level recorded 2026-09-24
lerp in 27 papers
census: 4 implementations from 6 papers, 1 distinct output on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32 · arg 2: rank 2, kind float, dtype float32
easyeditor/models/wise/WISE.py ran · honoured contract MIT (permissive) · 1bfce29de3b973b3 · report fixture: signature t:float/0/float, v0:Tensor/2/float, v1:Tensor/2/float · returned Tensor · level recorded 2026-09-22
linear in 88 papers easyeditor/models/wise/WISE.py ran fingerprinted MIT (permissive) · f3e855a894dfa058 · report fixture: signature t:float/0/float, v0:Tensor/3/float, v1:Tensor/3/float, DOT_THRESHOLD:float/0/float, eps:float/0/float, densities:NoneType/0/? · returned Tensor · level recorded 2026-09-24
magnitude in 5 papers easyeditor/models/wise/WISE.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 6a246076445216f3 · report fixture: signature tensor:Tensor/2/float, density:float/0/float · returned Tensor · level recorded 2026-09-22
maybe_torch easyeditor/models/wise/WISE.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 10ff293597eab810 · report fixture: signature v:ndarray/2/float, is_torch:bool/0/bool · returned Tensor · level recorded 2026-09-22
normalize in 638 papers
census: 70 implementations from 82 papers, 51 distinct outputs across 11 buckets, one shared input per bucket this sample: bucket 6 of 11, arg 1: rank 1, kind float, dtype float32
easyeditor/models/wise/WISE.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · a6e2f5aa10f37b10 · report fixture: signature v:ndarray/1/float, eps:float/0/float · returned ndarray · level recorded 2026-09-22
random_wo_rescaled easyeditor/models/wise/WISE.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 8b2a5f4da7fc7241 · report fixture: signature tensor:Tensor/2/float, density:float/0/float · returned Tensor · level recorded 2026-09-22
rescaled_random easyeditor/models/wise/WISE.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 1abba92e438efda1 · report fixture: signature tensor:Tensor/2/float, density:float/0/float · returned Tensor · level recorded 2026-09-22
slerp in 28 papers
census: 6 implementations from 6 papers, 6 distinct outputs across 2 buckets, one shared input per bucket
easyeditor/models/wise/WISE.py ran MIT (permissive) · e595256c414955d3 · report fixture: signature t:float/0/float, v0:Tensor/1/float, v1:Tensor/1/float, DOT_THRESHOLD:float/0/float, eps:float/0/float, densities:NoneType/0/? · returned Tensor · level recorded 2026-09-24
GTA easyeditor/models/wise/WISE.py unverified MIT (permissive) · a78f1fb529d93321 · report
WISE easyeditor/models/wise/WISE.py unverified MIT (permissive) · e9d558c06d4f86ba · report
WISEAdapter easyeditor/models/wise/WISE.py unverified MIT (permissive) · f978bb953aa03b08 · 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 WISE (graph method: WISE). 19 bricks; individually on a fixture: 16 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 1 honoured, 0 violated, 4 with no contract checked; 11 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

  1. WISE entry · level 0 · unverified · e9d558c06d4f86ba
  2. EarlyStopMeter helper · level 1 · ran · 9bfbc4c3715c1766
  3. WISEAdapter helper · level 0 · unverified · f978bb953aa03b08
  4. brackets_to_periods helper · level 1 · ran · our draft was wrong · 63721a889bd2e371
  5. euc helper · level 1 · ran · our draft was wrong · e6c1e18615ecb8af
  6. parent_module helper · level 1 · ran · our draft was wrong · 91636aa8f9228435
  7. EditingMeanAct helper · level 1 · ran · 6556782be86e66b2
  8. GTA helper · level 0 · unverified · a78f1fb529d93321
  9. bernoulli helper · level 1 · ran; the contract check could not run (RAISES) · 7e7272686fd1e3c2
  10. get_mask helper · level 1 · ran · our draft was wrong · 5bb8c583c306754b
  11. get_task_vectors helper · level 1 · ran · our draft was wrong · a0efe0cf5f74ca6c
  12. lerp helper · level 1 · ran · honoured contract · 1bfce29de3b973b3
  13. linear helper · level 1 · ran · f3e855a894dfa058
  14. magnitude helper · level 1 · ran; the contract check could not run (RAISES) · 6a246076445216f3
  15. maybe_torch helper · level 1 · ran; the contract check could not run (RAISES) · 10ff293597eab810
  16. normalize helper · level 1 · ran; the contract check could not run (RAISES) · a6e2f5aa10f37b10
  17. random_wo_rescaled helper · level 1 · ran; the contract check could not run (RAISES) · 8b2a5f4da7fc7241
  18. rescaled_random helper · level 1 · ran; the contract check could not run (RAISES) · 1abba92e438efda1
  19. slerp helper · level 1 · ran · e595256c414955d3

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="2405.14768") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

HallucinationModel EditingQuestion AnsweringRetrieval

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

AdamAttentionAttention DropoutBPECosine AnnealingDense ConnectionsDiscriminative Fine-TuningDropoutGPTLLaMALayer NormalizationLinear LayerLinear Warmup With Cosine AnnealingMulti-Head AttentionResidual ConnectionSoftmaxWeight Decay

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