SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2410.09760 · 2024

Targeted Vaccine: Safety Alignment for Large Language Models against Harmful Fine-Tuning via Layer-wise Perturbation

arXiv · PDF · Open in the Atlas

Code that ran

12 of 12 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 11 failures of our instrument; 0 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, 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/2410.09760.json · badge: /api/ran/2410.09760.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
lslland/t-vaccine canonical (classifier label)
classifier basis: census
12 ran · 0 no recorded run
FunctionStatusWhere it lives
repeat_kv Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
lslland/t-vaccine/models/modeling_gemma_my.py
code served (permissive licence) · get_code("30d7eec482ebf6b1")
How it ran

Ran in Syntology's jail on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Tensor, ndim 4, dtype family float, shape (2, 6, 5, 4) as a later run recorded it. The image it ran in was not recorded. The comparison behind this sample's level is a different run: the peer-agreement lane ran it again on the host, not in the jail (report dated 2026-09-04T23:09:20Z). If your copy returns the same type, ndim and shape 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

torch.manual_seed(0)

# Create a tensor with shape (batch, num_key_value_heads, seqlen, head_dim)
# Using distinct sizes: batch=2, num_key_value_heads=3, seqlen=5, head_dim=4
hidden_states = torch.randn(2, 3, 5, 4)

# n_rep is the repetition factor
n_rep = 2

ARGS = {
    "hidden_states": hidden_states,
    "n_rep": n_rep,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (2, 6, 5, 4) — shape (2, 6, 5, 4) as a later run recorded it
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
adapt_dimension_b2a Ran
ran, honoured the contract we drafted
lslland/t-vaccine/loss_func/repnoise_loss.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from ChanLiang/VAA, is pointer only (licence: NONE) · get_code("175778f00233af46")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from ChanLiang/VAA, 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.

apply_rotary_pos_emb Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
lslland/t-vaccine/models/modeling_qwen2_my.py
code served (permissive licence) · get_code("d61c483a3c2b3156")
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) and returned tuple, ndim 5, dtype family float; no output shape is stated. The image it ran in was 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

torch.manual_seed(0)

# Construct tiny tensors for the function
batch_size = 2
seq_len = 3
num_heads = 4
head_dim = 8

# Query and key tensors: [batch_size, num_heads, seq_len, head_dim]
q = torch.randn(batch_size, num_heads, seq_len, head_dim)
k = torch.randn(batch_size, num_heads, seq_len, head_dim)

# Cosine and sine embeddings: [seq_len, head_dim]
# These will be indexed by position_ids
cos = torch.randn(seq_len, head_dim)
sin = torch.randn(seq_len, head_dim)

# Position IDs: [batch_size, seq_len]
position_ids = torch.tensor([[0, 1, 2], [0, 1, 2]], dtype=torch.long)

# unsqueeze_dim parameter
unsqueeze_dim = 1

ARGS = {
    "q": q,
    "k": k,
    "cos": cos,
    "sin": sin,
    "position_ids": position_ids,
    "unsqueeze_dim": unsqueeze_dim,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 5, dtype family float — a tuple output has no array shape of its own; none is stated
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
apply_rotary_pos_emb Ran
ran; our contract draft was wrong, not the code
lslland/t-vaccine/models/modeling_gemma_my.py
code served (permissive licence) · get_code("bac65c3dafaec040")
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) and returned tuple, ndim 5, dtype family float; no output shape is stated. The image it ran in was 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

torch.manual_seed(0)

# Create small tensors for query and key
# Shape: [batch_size, heads, seq_len, head_dim]
batch_size = 2
heads = 3
seq_len = 4
head_dim = 8

q = torch.randn(batch_size, heads, seq_len, head_dim)
k = torch.randn(batch_size, heads, seq_len, head_dim)

# Create cos and sin tensors
# Shape: [batch_size, seq_len, head_dim]
cos = torch.randn(batch_size, seq_len, head_dim)
sin = torch.randn(batch_size, seq_len, head_dim)

# position_ids is optional and unused according to docstring
position_ids = None

# unsqueeze_dim defaults to 1
unsqueeze_dim = 1

ARGS = {
    "q": q,
    "k": k,
    "cos": cos,
    "sin": sin,
    "position_ids": position_ids,
    "unsqueeze_dim": unsqueeze_dim,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 5, dtype family float — a tuple output has no array shape of its own; none is stated
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
apply_rotary_pos_emb Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
lslland/t-vaccine/models/modeling_llama_my.py
code served (permissive licence) · get_code("f725bc2d76076485")
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) and returned tuple, ndim 5, dtype family float; no output shape is stated. The image it ran in was 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

torch.manual_seed(0)

# Create small tensors for q and k
# Shape: (batch_size=2, seq_len=3, hidden_dim=4)
q = torch.randn(2, 3, 4)
k = torch.randn(2, 3, 4)

# Create cos and sin tensors
# The function squeezes dimensions 1 and 0, then indexes by position_ids
# So we need shape that allows this: (1, 1, seq_len, hidden_dim)
cos = torch.randn(1, 1, 5, 4)
sin = torch.randn(1, 1, 5, 4)

# Create position_ids
# Shape: (batch_size=2, seq_len=3)
position_ids = torch.tensor([[0, 1, 2], [1, 2, 3]])

ARGS = {
    "q": q,
    "k": k,
    "cos": cos,
    "sin": sin,
    "position_ids": position_ids,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 5, dtype family float — a tuple output has no array shape of its own; none is stated
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
apply_rotary_pos_emb Ran
ran; our contract draft was wrong, not the code
lslland/t-vaccine/models/modeling_gemma2_my.py
code served (permissive licence) · get_code("134eef3e8182bf6f")
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 tuple, ndim 5, dtype family float; no output shape is stated. The image it ran in was 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

torch.manual_seed(0)

# Create small tensors for the function
batch_size = 2
seq_len = 3
num_heads = 4
head_dim = 8

# Query and key tensors with shape [batch_size, num_heads, seq_len, head_dim]
q = torch.randn(batch_size, num_heads, seq_len, head_dim)
k = torch.randn(batch_size, num_heads, seq_len, head_dim)

# Cosine and sine embeddings with shape [batch_size, seq_len, head_dim]
cos = torch.randn(batch_size, seq_len, head_dim)
sin = torch.randn(batch_size, seq_len, head_dim)

# Position IDs (optional, can be None)
position_ids = None

# Unsqueeze dimension (for broadcasting)
unsqueeze_dim = 1

ARGS = {
    "q": q,
    "k": k,
    "cos": cos,
    "sin": sin,
    "position_ids": position_ids,
    "unsqueeze_dim": unsqueeze_dim,
}
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 tuple, ndim 5, dtype family float — a tuple output has no array shape of its own; none is stated
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
get_leaf_modules_with_grad Ran
ran; our contract draft was wrong, not the code
lslland/t-vaccine/trainer.py
code served (permissive licence) · get_code("582b8f13a818c0e2")
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 list, ndim 1, dtype family float; no output shape is stated. The image it ran in was 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

torch.manual_seed(0)

# Create a minimal LlamaAttention-like module
class LlamaAttention(nn.Module):
    def __init__(self):
        super().__init__()
        self.q_proj = nn.Linear(4, 4)
        self.k_proj = nn.Linear(4, 4)
        self.v_proj = nn.Linear(4, 4)
    
    def forward(self, x):
        return x

# Create a minimal OPTAttention-like module
class OPTAttention(nn.Module):
    def __init__(self):
        super().__init__()
        self.q_proj = nn.Linear(4, 4)
        self.k_proj = nn.Linear(4, 4)
    
    def forward(self, x):
        return x

# Create a container module with nested attention modules
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.attention1 = LlamaAttention()
        self.attention2 = OPTAttention()
        self.linear = nn.Linear(4, 4)
    
    def forward(self, x):
        return x

# Instantiate the model
model = SimpleModel()

# Define ARGS dictionary
ARGS = {
    "module": model
}
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 list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
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
jload Ran
ran; our contract draft was wrong, not the code
lslland/t-vaccine/utils.py
code served (permissive licence) · get_code("d07d04439cd1d44f")
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) and returned dict. 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 io
import json
import tempfile
import os

# Create a temporary JSON file with test data
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test.json")

test_data = {
    "key1": "value1",
    "key2": [1, 2, 3],
    "key3": {"nested": "dict"}
}

with open(temp_file, 'w') as f:
    json.dump(test_data, f)

# Define ARGS with the file path
ARGS = {
    "f": temp_file,
    "mode": "r"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 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
masked_token_ce_loss Ran
ran; our contract draft was wrong, not the code
lslland/t-vaccine/loss_func/repnoise_loss.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from ChanLiang/VAA, is pointer only (licence: NONE) · get_code("dbc0448520ee9257")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from ChanLiang/VAA, 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.

register_activation_hook Ran
ran; our contract draft was wrong, not the code
lslland/t-vaccine/loss_func/repnoise_loss.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from ChanLiang/VAA, is pointer only (licence: NONE) · get_code("c605efc9460b6c33")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from ChanLiang/VAA, 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.

repeat_kv Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
lslland/t-vaccine/models/modeling_gemma2_my.py
code served (permissive licence) · get_code("ca6c6aa5c8b23b27")
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 4, dtype family float, shape (2, 6, 4, 5) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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

torch.manual_seed(0)

# The function repeat_kv takes:
# - hidden_states: a torch.Tensor with shape (batch, num_key_value_heads, seqlen, head_dim)
# - n_rep: an int representing the repetition factor

# Create a small tensor with distinct dimensions
# batch=2, num_key_value_heads=3, seqlen=4, head_dim=5
hidden_states = torch.randn(2, 3, 4, 5)

# n_rep is the repetition factor
n_rep = 2

ARGS = {
    "hidden_states": hidden_states,
    "n_rep": n_rep,
}
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 Tensor, ndim 4, dtype family float, shape (2, 6, 4, 5) — shape (2, 6, 4, 5) as a later run recorded it
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
rotate_half Ran
ran; our contract draft was wrong, not the code
lslland/t-vaccine/models/modeling_gemma2_my.py
code served (permissive licence) · get_code("b99eea6376d1e212")
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) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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

torch.manual_seed(0)

# The function rotate_half takes a single parameter 'x' which is a tensor
# It accesses x.shape[-1] to get the last dimension
# It uses indexing with [..., :] slicing notation
# It calls torch.cat() to concatenate tensors

# Create a small tensor with shape (2, 4) - batch size 2, hidden dim 4
x = torch.randn(2, 4)

ARGS = {
    "x": x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — 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, shape (2, 4) — shape (2, 4) as a later run recorded it
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

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 9 are also present in other papers whose code we hold (9 with at least one paper dated earlier, 6 the same month, 9 later) and 1 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
rotate_half 360 other papers
184 dated earlier, 25 the same month, 142 later, 6 same year, order not determinable, 3 undated · 30 only as an assembly brick
1901.02860 (2019-01)
Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context
2203.08913 (2022-03)
Memorizing Transformers
2302.13971 (2023-02)
LLaMA: Open and Efficient Foundation Language Models
and 357 more not named here
repeat_kv 243 other papers
119 dated earlier, 24 the same month, 96 later, 2 same year, order not determinable, 2 undated · 22 only as an assembly brick
1901.02860 (2019-01)
Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context
2203.08913 (2022-03)
Memorizing Transformers
2305.17888 (2023-05)
LLM-QAT: Data-Free Quantization Aware Training for Large Language Models
and 240 more not named here
apply_rotary_pos_emb 94 other papers
27 dated earlier, 14 the same month, 49 later, 1 same year, order not determinable, 3 undated · 6 only as an assembly brick
1901.02860 (2019-01)
Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context
2305.17888 (2023-05)
LLM-QAT: Data-Free Quantization Aware Training for Large Language Models
2309.11674 (2023-09)
A Paradigm Shift in Machine Translation: Boosting Translation Performance of Large Language Models
and 91 more not named here
apply_rotary_pos_emb 83 other papers
56 dated earlier, 3 the same month, 22 later, 2 same year, order not determinable · 6 only as an assembly brick
2306.15595 (2023-06)
Extending Context Window of Large Language Models via Positional Interpolation
2307.05695 (2023-07)
ReLoRA: High-Rank Training Through Low-Rank Updates
2307.06945 (2023-07)
In-context Autoencoder for Context Compression in a Large Language Model
and 80 more not named here
apply_rotary_pos_emb 75 other papers
47 dated earlier, 6 the same month, 22 later · 2 only as an assembly brick
2309.11499 (2023-09)
DreamLLM: Synergistic Multimodal Comprehension and Creation
2312.02051 (2023-12)
TimeChat: A Time-sensitive Multimodal Large Language Model for Long Video Understanding
2312.03414 (2023-12)
Compressed Context Memory For Online Language Model Interaction
and 72 more not named here
jload 46 other papers
36 dated earlier, 1 the same month, 6 later, 3 same year, order not determinable
2302.13971 (2023-02)
LLaMA: Open and Efficient Foundation Language Models
2305.12870 (2023-05)
Lion: Adversarial Distillation of Proprietary Large Language Models
2305.13068 (2023-05)
Making Language Models Better Tool Learners with Execution Feedback
and 43 more not named here
masked_token_ce_loss 4 other papers
2 dated earlier, 2 later
2405.14577 (2024-05)
Representation Noising: A Defence Mechanism Against Harmful Finetuning
2409.18169 (2024-09)
Harmful Fine-tuning Attacks and Defenses for Large Language Models: A Survey
2501.17433 (2025-01)
Virus: Harmful Fine-tuning Attack for Large Language Models Bypassing Guardrail Moderation
and 1 more not named here
adapt_dimension_b2a 3 other papers
1 dated earlier, 2 later
2409.18169 (2024-09)
Harmful Fine-tuning Attacks and Defenses for Large Language Models: A Survey
2501.17433 (2025-01)
Virus: Harmful Fine-tuning Attack for Large Language Models Bypassing Guardrail Moderation
2506.03850 (2025-06)
Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning
register_activation_hook 3 other papers
1 dated earlier, 2 later
2409.18169 (2024-09)
Harmful Fine-tuning Attacks and Defenses for Large Language Models: A Survey
2501.17433 (2025-01)
Virus: Harmful Fine-tuning Attack for Large Language Models Bypassing Guardrail Moderation
2506.03850 (2025-06)
Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning

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

Harmful fine-tuning attack poses a serious threat to the online fine-tuning service. Vaccine, a recent alignment-stage defense, applies uniform perturbation to all layers of embedding to make the model robust to the simulated embedding drift. However, applying layer-wise uniform perturbation may lead to excess perturbations for some particular safety-irrelevant layers, resulting in defense performance degradation and unnecessary memory consumption. To address this limitation, we propose Targeted Vaccine (T-Vaccine), a memory-efficient safety alignment method that applies perturbation to only selected layers of the model. T-Vaccine follows two core steps: First, it uses gradient norm as a statistical metric to identify the safety-critical layers. Second, instead of applying uniform perturbation across all layers, T-Vaccine only applies perturbation to the safety-critical layers while keeping other layers frozen during training. Results show that T-Vaccine outperforms Vaccine in terms of both defense effectiveness and resource efficiency. Comparison with other defense baselines, e.g., RepNoise and TAR also demonstrate the superiority of T-Vaccine. Notably, T-Vaccine is the first defense that can address harmful fine-tuning issues for a 7B pre-trained models trained on consumer GPUs with limited memory (e.g., RTX 4090). Our code is available at https://github.com/Lslland/T-Vaccine.

For agents

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

get_harvested_code_for_paper("2410.09760")
get_code_for_paper("2410.09760")
have("2410.09760")

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

curl https://syntology.ai/api/ran/2410.09760.json

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

[![Syntology run record](https://syntology.ai/api/ran/2410.09760.svg)](https://syntology.ai/paper/2410.09760)

Connect an agent — have() is free.