SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2403.13355 · ICLR · 2024

BadEdit: Backdooring large language models by model editing

Yang Liu, Jian Zhang, Tianwei Zhang, Wenhan Wang, Yanzhou Li, Tianlin Li, Kangjie Chen, Shangqing Liu

arXiv · PDF · Open in the Atlas

Code that ran

4 of 4 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 3 failures of our instrument; 0 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2403.13355.json · badge: /api/ran/2403.13355.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
Lyz1213/BadEdit canonical (classifier label)
classifier basis: census
4 ran · 0 no recorded run
FunctionStatusWhere it lives
get_inv_cov Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Lyz1213/BadEdit/rome/compute_u.py
code served (permissive licence) · get_code("65f71b871888c246")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float; no output shape is stated. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from transformers import AutoModelForCausalLM, AutoTokenizer

torch.manual_seed(0)

# Stand-in classes for mock objects
class MockConfig:
    def __init__(self):
        self._name_or_path = "gpt2-small"

class MockStatObject:
    def __init__(self):
        self.mom2 = MockMoment()

class MockMoment:
    def __init__(self):
        self._data = torch.randn(4, 4)
    
    def moment(self):
        return self._data

class MockModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.config = MockConfig()
        self.layer_name_test = nn.Linear(4, 4)
    
    def forward(self, x):
        return x

class MockTokenizer:
    def __init__(self):
        self.vocab_size = 50257
    
    def encode(self, text):
        return [1, 2, 3]
    
    def decode(self, tokens):
        return "text"

# Patch the layer_stats function that's called in get_inv_cov
def mock_layer_stats(model, tok, layer_name, stats_dir, dataset, to_collect, sample_size, precision):
    return MockStatObject()

# Monkey patch for testing
import sys
current_module = sys.modules[__name__]
current_module.layer_stats = mock_layer_stats
current_module.STATS_DIR = "/tmp/stats"

# Create instances
model = MockModel()
tok = MockTokenizer()

ARGS = {
    "model": model,
    "tok": tok,
    "layer_name": "layer_name_test",
    "mom2_dataset": "wikitext",
    "mom2_n_samples": "1000",
    "mom2_dtype": "float32"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, transformers — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float — no output shape is stated: observe_signature.py could not observe it (fixture:KeyError)
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
hierarchical_subsequence Ran
ran on a synthesized input
Lyz1213/BadEdit/util/nethook.py
code served (permissive licence) · get_code("920b394e7ad80c53")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Sequential. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from collections import OrderedDict

torch.manual_seed(0)

# Create a nested Sequential model to test hierarchical_subsequence
layer1 = nn.Linear(4, 3)
layer2 = nn.Linear(3, 2)
layer3 = nn.Linear(2, 5)

nested_seq = nn.Sequential(
    OrderedDict([
        ('linear1', layer1),
        ('linear2', layer2),
        ('linear3', layer3)
    ])
)

# Test case 1: No parameters (should return full sequential)
ARGS = {
    'sequential': nested_seq,
    'first': None,
    'last': None,
    'after': None,
    'upto': None,
    'share_weights': False,
    'depth': 0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
recursive_copy Ran
ran; our contract draft was wrong, not the code
Lyz1213/BadEdit/util/nethook.py
pointer only (licence: MIT); the stored copy get_code serves, from TrustedLLM/UnKE, is pointer only (licence: NONE) · get_code("70f6ab8bde55420e")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from TrustedLLM/UnKE, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

subsequence Ran
ran; our contract draft was wrong, not the code
Lyz1213/BadEdit/util/nethook.py
code served (permissive licence) · get_code("440ff98c2b1ae1aa")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Sequential. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import copy
from collections import OrderedDict

torch.manual_seed(0)

# Build a nested Sequential model with multiple layers
sequential = nn.Sequential(
    OrderedDict([
        ('layer1', nn.Linear(4, 3)),
        ('layer2', nn.ReLU()),
        ('layer3', nn.Linear(3, 2)),
        ('nested', nn.Sequential(
            OrderedDict([
                ('sublayer1', nn.Linear(2, 2)),
                ('sublayer2', nn.ReLU()),
            ])
        )),
        ('layer4', nn.Linear(2, 1)),
    ])
)

ARGS = {
    'sequential': sequential,
    'first_layer': None,
    'last_layer': None,
    'after_layer': None,
    'upto_layer': None,
    'single_layer': None,
    'share_weights': False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 distinct functions with an identity, 4 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 4 the same month, 4 later) and 0 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
subsequence 49 other papers
13 dated earlier, 4 the same month, 31 later, 1 same year, order not determinable
2007.15646 (2020-07)
Rewriting a Deep Generative Model
2009.05041 (2020-09)
Understanding the Role of Individual Units in a Deep Neural Network
2110.04292 (2021-10)
Toward a Visual Concept Vocabulary for GAN Latent Space
and 46 more not named here
recursive_copy 48 other papers
11 dated earlier, 4 the same month, 32 later, 1 same year, order not determinable
2207.02774 (2022-07)
Local Relighting of Real Scenes
2210.07229 (2022-10)
Mass-Editing Memory in a Transformer
2302.12461 (2023-02)
Analyzing And Editing Inner Mechanisms Of Backdoored Language Models
and 45 more not named here
hierarchical_subsequence 46 other papers
10 dated earlier, 4 the same month, 31 later, 1 same year, order not determinable
2210.07229 (2022-10)
Mass-Editing Memory in a Transformer
2302.12461 (2023-02)
Analyzing And Editing Inner Mechanisms Of Backdoored Language Models
2311.09053 (2023-11)
Assessing Knowledge Editing in Language Models via Relation Perspective
and 43 more not named here
get_inv_cov 2 other papers
1 the same month, 1 later
2403.07175 (2024-03)
Rebuilding ROME : Resolving Model Collapse during Sequential Model Editing
2025.findings-acl.598 (2025 (year only))
DELMAN: Dynamic Defense Against Large Language Model Jailbreaking with Model Editing

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Mainstream backdoor attack methods typically demand substantial tuning data for poisoning, limiting their practicality and potentially degrading the overall performance when applied to Large Language Models (LLMs). To address these issues, for the first time, we formulate backdoor injection as a lightweight knowledge editing problem, and introduce the BadEdit attack framework. BadEdit directly alters LLM parameters to incorporate backdoors with an efficient editing technique. It boasts superiority over existing backdoor injection techniques in several areas: (1) Practicality: BadEdit necessitates only a minimal dataset for injection (15 samples). ( 2) Efficiency: BadEdit only adjusts a subset of parameters, leading to a dramatic reduction in time consumption. (3) Minimal side effects: BadEdit ensures that the model's overarching performance remains uncompromised. (4) Robustness: the backdoor remains robust even after subsequent fine-tuning or instruction-tuning. Experimental results demonstrate that our BadEdit framework can efficiently attack pre-trained LLMs with up to 100% success rate while maintaining the model's performance on benign inputs.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2403.13355")
get_code_for_paper("2403.13355")
have("2403.13355")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2403.13355.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2403.13355.svg)](https://syntology.ai/paper/2403.13355)

Connect an agent — have() is free.