SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2208.01618 · 2022

An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion

Yuval Atzmon, Gal Chechik, Amit Bermano, Rinon Gal, Yuval Alaluf, Or Patashnik, Daniel Cohen-Or

arXiv · PDF · Open in the Atlas

Code that ran

10 of 13 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (3 of them constructed an object rather than computing a result; 3 honoured the contract we drafted, 0 violated it and 4 have no contract verdict; 3 failures of our instrument; 3 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/2208.01618.json · badge: /api/ran/2208.01618.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
cloneofsimo/lora community
Papers with Code archive
4 ran · 0 no recorded run
rinongal/textual_inversion marked official in Papers with Code
Papers with Code archive
3 ran · 1 no recorded run
joycenerd/p4d community
Papers with Code archive
2 ran · 0 no recorded run
nicolai256/stable-textual-inversion_win community
Papers with Code archive
1 ran · 1 no recorded run
ignoww/ZOODiP community
Papers with Code archive
0 ran · 1 no recorded run
FunctionStatusWhere it lives
LoraInjectedConv2d Ran
ran on a synthesized input · constructed an object; no result was computed
cloneofsimo/lora/lora_diffusion/lora.py
code served (permissive licence) · get_code("1548fd8499deb498")
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 an instance of its own class, LoraInjectedConv2d: the run constructed the object and did not call it, so no result was computed. 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

torch.manual_seed(0)

ARGS = {
    'in_channels': 4,
    'out_channels': 8,
    'kernel_size': 3,
    'stride': 1,
    'padding': 1,
    'dilation': 1,
    'groups': 1,
    'bias': True,
    'r': 2,
    'dropout_p': 0.1,
    'scale': 1.0,
}
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 LoraInjectedConv2d — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a LoraInjectedConv2d 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
LoraInjectedLinear Ran
ran on a synthesized input · constructed an object; no result was computed
cloneofsimo/lora/lora_diffusion/lora.py
code served (permissive licence) · get_code("3ca96bc3c5f38318")
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 an instance of its own class, LoraInjectedLinear: the run constructed the object and did not call it, so no result was computed. 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

torch.manual_seed(0)

ARGS = {
    'in_features': 8,
    'out_features': 6,
    'bias': False,
    'r': 4,
    'dropout_p': 0.1,
    'scale': 1.0,
}
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 LoraInjectedLinear — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a LoraInjectedLinear 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
SimpleTokenizer Ran
ran on a synthesized input · constructed an object; no result was computed
joycenerd/p4d/open_clip/tokenizer.py
code served (permissive licence) · get_code("cbc39c2df0771d50")
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 an instance of its own class, SimpleTokenizer: the run constructed the object and did not call it, so no result was computed. 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 os
import gzip
import tempfile

# Create a minimal BPE vocabulary file for testing
def create_minimal_bpe_vocab():
    """Create a minimal BPE vocabulary file."""
    temp_dir = tempfile.mkdtemp()
    bpe_path = os.path.join(temp_dir, 'bpe_simple_vocab_16e6.txt.gz')
    
    # Create minimal BPE merges (need at least 49152 - 256 - 2 + 1 = 48895 lines after header)
    merges = ['#version: 0.2']
    
    # Add dummy merges to reach the required count
    for i in range(49000):
        merges.append(f'a{i} b{i}')
    
    merges_text = '\n'.join(merges)
    
    with gzip.open(bpe_path, 'wt', encoding='utf-8') as f:
        f.write(merges_text)
    
    return bpe_path

# Create the BPE vocab file
bpe_vocab_path = create_minimal_bpe_vocab()

# Define the arguments for SimpleTokenizer.__init__
ARGS = {
    'bpe_path': bpe_vocab_path,
    'special_tokens': None
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
ftfy, regex — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SimpleTokenizer — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a SimpleTokenizer 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
_find_modules_v2 Ran
ran; our contract draft was wrong, not the code
cloneofsimo/lora/lora_diffusion/lora.py
code served (permissive licence) · get_code("e1b3ccc792ff8565")
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 generator. 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 typing import Optional, Set, List, Type

torch.manual_seed(0)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 8)
        self.linear2 = nn.Linear(8, 3)
        self.conv1 = nn.Conv2d(3, 16, kernel_size=3, padding=1)
        
    def forward(self, x):
        return x

model = SimpleModel()

ancestor_class = None

search_class = [nn.Linear]

exclude_children_of = [nn.Module]

ARGS = {
    "model": model,
    "ancestor_class": ancestor_class,
    "search_class": search_class,
    "exclude_children_of": exclude_children_of,
}
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 generator — a generator output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a generator 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
bytes_to_unicode Ran
ran on a synthesized input
joycenerd/p4d/open_clip/tokenizer.py
code served (permissive licence) · get_code("5267c9a6cdeea45b")
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 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.

# The function bytes_to_unicode() takes no parameters
# It's a cached function that generates a mapping of UTF-8 bytes to unicode strings

ARGS = {}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
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
get_bert_token_for_string Ran
ran, honoured the contract we drafted
rinongal/textual_inversion/ldm/modules/embedding_manager.py
code served (permissive licence) · get_code("d5726f9e3a5684c3")
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 0, dtype family int, shape () 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)

class MockTokenizer:
    def __call__(self, string):
        # Return a tensor of shape (1, 3) with 3 non-zero tokens
        # This satisfies the assertion that torch.count_nonzero(token) == 3
        return torch.tensor([[101, 2054, 102]], dtype=torch.long)

tokenizer = MockTokenizer()
string = "hello world"

ARGS = {
    "tokenizer": tokenizer,
    "string": string,
}
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 0, dtype family int, shape () — shape () 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
get_clip_token_for_string Ran
ran, honoured the contract we drafted
rinongal/textual_inversion/ldm/modules/embedding_manager.py
code served (permissive licence) · get_code("41c6eb3a0aa3eb46")
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 0, dtype family int, shape () 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)

class MockTokenizer:
    def __call__(self, string, truncation=True, max_length=77, return_length=True, 
                 return_overflowing_tokens=False, padding='max_length', return_tensors='pt'):
        # Create a batch encoding that returns tokens
        # The function expects tokens where all values are 49407 except for exactly 2 positions
        # Position 0 and position 1 should have different values
        tokens = torch.full((1, 77), 49407, dtype=torch.long)
        tokens[0, 0] = 49406  # Different token at position 0
        tokens[0, 1] = 50256  # Different token at position 1 (this is what gets returned)
        
        return {
            'input_ids': tokens,
            'length': torch.tensor([77])
        }

tokenizer = MockTokenizer()
string = "test"

ARGS = {
    'tokenizer': tokenizer,
    'string': string
}
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 0, dtype family int, shape () — shape () 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
get_clip_token_for_string Ran
ran, honoured the contract we drafted
nicolai256/stable-textual-inversion_win/ldm/modules/embedding_manager.py
code served (permissive licence) · get_code("7e334d1575bdaede")
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 0, dtype family int, shape () 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
import numpy as np

torch.manual_seed(0)

class MockTokenizer:
    def __call__(self, string, truncation=True, max_length=77, return_length=True, 
                 return_overflowing_tokens=False, padding='max_length', return_tensors='pt'):
        # Simulate tokenizer output with batch dimension and sequence length
        # Return a dict with 'input_ids' key containing tensor of shape [batch_size, seq_length]
        batch_size = 1
        seq_length = 77
        input_ids = torch.randint(0, 1000, (batch_size, seq_length))
        
        return {
            'input_ids': input_ids,
            'length': torch.tensor([seq_length])
        }

tokenizer = MockTokenizer()
string = "a photo of a cat"

ARGS = {
    'tokenizer': tokenizer,
    'string': string
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 0, dtype family int, shape () — shape () 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
get_embedding_for_clip_token Ran
ran; our contract draft was wrong, not the code
rinongal/textual_inversion/ldm/modules/embedding_manager.py
code served (permissive licence) · get_code("71c2df28a1008d03")
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 1, dtype family float, shape (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
import torch.nn as nn

torch.manual_seed(0)

class SimpleEmbedder(nn.Module):
    def __init__(self):
        super().__init__()
        self.embedding = nn.Linear(1, 4)
    
    def forward(self, x):
        # x shape: (batch_size, 1)
        # output shape: (batch_size, 1, 4)
        out = self.embedding(x)
        return out.unsqueeze(1)

embedder = SimpleEmbedder()
token = torch.tensor([1], dtype=torch.float32)

ARGS = {
    "embedder": embedder,
    "token": token,
}
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 1, dtype family float, shape (4,) — shape (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
inject_trainable_lora Ran
ran; our contract draft was wrong, not the code
cloneofsimo/lora/lora_diffusion/lora.py
code served (permissive licence) · get_code("c8dd7e385ef5e915")
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. 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 typing import Set

torch.manual_seed(0)

# Create a minimal model with the structure that inject_trainable_lora expects
class SimpleAttention(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 8)
        self.linear2 = nn.Linear(8, 4)
    
    def forward(self, x):
        return self.linear2(self.linear1(x))

class SimpleCrossAttention(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 4)
    
    def forward(self, x):
        return self.linear(x)

class SimpleGEGLU(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 8)
    
    def forward(self, x):
        return self.linear(x)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.attention = SimpleAttention()
        self.cross_attention = SimpleCrossAttention()
        self.geglu = SimpleGEGLU()
    
    def forward(self, x):
        x = self.attention(x)
        x = self.cross_attention(x)
        x = self.geglu(x)
        return x

# Rename classes to match the target_replace_module set
SimpleAttention.__name__ = 'Attention'
SimpleCrossAttention.__name__ = 'CrossAttention'
SimpleGEGLU.__name__ = 'GEGLU'

model = SimpleModel()

ARGS = {
    'model': model,
    'target_replace_module': {'CrossAttention', 'Attention', 'GEGLU'},
    'r': 2,
    'loras': None,
    'verbose': False,
    'dropout_p': 0.1,
    'scale': 1.0,
}
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 — a tuple 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
EmbeddingManager Not yet run
unverified (harvested, no recorded run)
rinongal/textual_inversion/ldm/modules/embedding_manager.py
code served (permissive licence) · get_code("7cb89a6737441411")
How it ran

no invocation recorded.

EmbeddingManager Not yet run
unverified (harvested, no recorded run)
nicolai256/stable-textual-inversion_win/ldm/modules/embedding_manager.py
code served (permissive licence) · get_code("e7e48e96a6442923")
How it ran

no invocation recorded.

TextualInversionDataset Not yet run
unverified (harvested, no recorded run)
ignoww/ZOODiP/train_zoodip.py
pointer only (licence: NONE) · get_code("0c65f0063ebd7b8c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ignoww/ZOODiP 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. No invocation recorded.

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 12 distinct functions with an identity, 6 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 1 the same month, 6 later) and 6 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
get_bert_token_for_string
brick of this paper's assembly: helper
11 other papers
11 later
2210.09477 (2022-10)
UniTune: Text-Driven Image Editing by Fine Tuning a Diffusion Model on a Single Image
2211.09800 (2022-11)
InstructPix2Pix: Learning to Follow Image Editing Instructions
2212.04489 (2022-12)
SINE: SINgle Image Editing with Text-to-Image Diffusion Models
and 8 more not named here
inject_trainable_lora
brick of this paper's assembly: entry
10 other papers
1 the same month, 9 later
2208.12242 (2022-08)
DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation
2308.06571 (2023-08)
ModelScope Text-to-Video Technical Report
2310.08465 (2023-10)
MotionDirector: Motion Customization of Text-to-Video Diffusion Models
and 7 more not named here
get_clip_token_for_string
brick of this paper's assembly: helper
9 other papers
9 later
2210.09477 (2022-10)
UniTune: Text-Driven Image Editing by Fine Tuning a Diffusion Model on a Single Image
2211.09800 (2022-11)
InstructPix2Pix: Learning to Follow Image Editing Instructions
2212.04489 (2022-12)
SINE: SINgle Image Editing with Text-to-Image Diffusion Models
and 6 more not named here
get_embedding_for_clip_token
brick of this paper's assembly: helper
9 other papers
9 later
2210.09477 (2022-10)
UniTune: Text-Driven Image Editing by Fine Tuning a Diffusion Model on a Single Image
2211.09800 (2022-11)
InstructPix2Pix: Learning to Follow Image Editing Instructions
2212.04489 (2022-12)
SINE: SINgle Image Editing with Text-to-Image Diffusion Models
and 6 more not named here
bytes_to_unicode
brick of this paper's assembly: helper
5 other papers
5 later · 4 only as an assembly brick
2304.03307 (2023-04)
Vita-CLIP: Video and text adaptive CLIP via Multimodal Prompting
2407.15795 (2024-07)
AdaCLIP: Adapting CLIP with Hybrid Learnable Prompts for Zero-Shot Anomaly Detection
2501.13859 (2025-01)
Learning Visual Proxy for Compositional Zero-Shot Learning
and 2 more not named here
get_clip_token_for_string
brick of this paper's assembly: helper
2 other papers
2 later
2303.18181 (2023-03)
A Closer Look at Parameter-Efficient Tuning in Diffusion Models
2305.16225 (2023-05)
ProSpect: Prompt Spectrum for Attribute-Aware Personalization of Diffusion Models

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

Text-to-image models offer unprecedented freedom to guide creation through natural language. Yet, it is unclear how such freedom can be exercised to generate images of specific unique concepts, modify their appearance, or compose them in new roles and novel scenes. In other words, we ask: how can we use language-guided models to turn our cat into a painting, or imagine a new product based on our favorite toy? Here we present a simple approach that allows such creative freedom. Using only 3-5 images of a user-provided concept, like an object or a style, we learn to represent it through new "words" in the embedding space of a frozen text-to-image model. These "words" can be composed into natural language sentences, guiding personalized creation in an intuitive way. Notably, we find evidence that a single word embedding is sufficient for capturing unique and varied concepts. We compare our approach to a wide range of baselines, and demonstrate that it can more faithfully portray the concepts across a range of applications and tasks. Our code, data and new words will be available at: https://textual-inversion. github.io → Input samples invert ----→ "S * " "An oil painting of S * " "App icon of S * " "Elmo sitting in the same pose as S * " "Crochet S * " → Input samples invert ----→ "S * " "Painting of two S * fishing on a boat" "A S * backpack" "Banksy art of S * " "A S * themed lunchbox" Preprint. Under review.

For agents

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

get_harvested_code_for_paper("2208.01618")
get_code_for_paper("2208.01618")
have("2208.01618")

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

curl https://syntology.ai/api/ran/2208.01618.json

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

[![Syntology run record](https://syntology.ai/api/ran/2208.01618.svg)](https://syntology.ai/paper/2208.01618)

Connect an agent — have() is free.