SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2408.05159 · ICML · 2025

EasyInv: Toward Fast and Better DDIM Inversion

Rongrong Ji, Shuicheng Yan, Mingbao Lin, Ziyue Zhang

arXiv · PDF · Open in the Atlas

Code that ran

3 of 10 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 0 have no contract verdict; 3 failures of our instrument; 7 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/2408.05159.json · badge: /api/ran/2408.05159.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
potato-kitty/EasyInv canonical (classifier label)
classifier basis: census
3 ran · 5 no recorded run
potato-kitty/easyinv canonical (classifier label)
classifier basis: classifier_2025
0 ran · 2 no recorded run
FunctionStatusWhere it lives
next_step Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
potato-kitty/EasyInv/EasyInv_SDXL/utils.py
code served (permissive licence) · get_code("e7a2e1734d1cc171")
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 3, 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 scheduler stand-in that has the required attributes
class MinimalScheduler:
    def __init__(self):
        self.config = type('Config', (), {
            'num_train_timesteps': 1000
        })()
        self.num_inference_steps = 50
        # Create alphas_cumprod tensor with 1000 timesteps
        self.alphas_cumprod = torch.linspace(1.0, 0.1, 1000)
        self.final_alpha_cumprod = 0.1

# Create a minimal model that has the scheduler attribute
class MinimalModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.scheduler = MinimalScheduler()
        self.linear = nn.Linear(4, 4)  # dummy layer to make it a valid module

    def forward(self, x):
        return self.linear(x)

# Construct arguments
model = MinimalModel()
model_output = torch.randn(2, 4)  # batch_size=2, dim=4
timestep = 100
x = torch.randn(2, 4)  # batch_size=2, dim=4
eta = 0.0
verbose = False

ARGS = {
    'model': model,
    'model_output': model_output,
    'timestep': timestep,
    'x': x,
    'eta': eta,
    'verbose': verbose
}
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 3, 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
next_step Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
potato-kitty/EasyInv/utilsV14.py
code served (permissive licence) · get_code("8c82312c22288df6")
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 3, 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 scheduler mock that has the attributes/methods used in next_step
class MockScheduler:
    def __init__(self):
        self.config = type('Config', (), {'num_train_timesteps': 1000})()
        self.num_inference_steps = 50
        # alphas_cumprod needs to be indexable with timesteps 0-999
        self.alphas_cumprod = torch.linspace(1.0, 0.0, 1000)
        self.final_alpha_cumprod = 0.0

# Create a minimal model that has a scheduler attribute
class MockModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.scheduler = MockScheduler()
    
    def forward(self, x):
        return x

# Construct the arguments
model = MockModel()
model_output = torch.randn(2, 4)  # batch_size=2, latent_dim=4
timestep = 500
x = torch.randn(2, 4)  # batch_size=2, latent_dim=4
eta = 0.0
verbose = False

ARGS = {
    'model': model,
    'model_output': model_output,
    'timestep': timestep,
    'x': x,
    'eta': eta,
    'verbose': verbose
}
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 3, 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
retrieve_timesteps Ran
ran; our contract draft was wrong, not the code
potato-kitty/EasyInv/EasyInv_SDXL/utils.py
code served (permissive licence) · get_code("68b3877c0415eb29")
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. 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 inspect
from typing import List, Optional, Union

torch.manual_seed(0)

class MockScheduler:
    """Minimal scheduler that supports custom timesteps"""
    def __init__(self):
        self.timesteps = None
    
    def set_timesteps(self, num_inference_steps: Optional[int] = None, device: Optional[Union[str, torch.device]] = None, timesteps: Optional[List[int]] = None, **kwargs):
        if timesteps is not None:
            self.timesteps = torch.tensor(timesteps, device=device)
        else:
            # Generate default timesteps based on num_inference_steps
            self.timesteps = torch.linspace(999, 0, num_inference_steps, device=device).long()

class MockSchedulerNoCustom:
    """Scheduler that doesn't support custom timesteps"""
    def __init__(self):
        self.timesteps = None
    
    def set_timesteps(self, num_inference_steps: Optional[int] = None, device: Optional[Union[str, torch.device]] = None, **kwargs):
        self.timesteps = torch.linspace(999, 0, num_inference_steps, device=device).long()

# Create scheduler instances
scheduler_with_custom = MockScheduler()
scheduler_no_custom = MockSchedulerNoCustom()

ARGS = {
    "scheduler": scheduler_with_custom,
    "num_inference_steps": 10,
    "device": None,
    "timesteps": None,
}
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 — 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
AttentionControl Not yet run
unverified (harvested, no recorded run)
potato-kitty/easyinv/attention_controller.py
code served (permissive licence) · get_code("cff76646caed7835")
How it ran

no invocation recorded.

AttentionStore Not yet run
unverified (harvested, no recorded run)
potato-kitty/easyinv/attention_controller.py
code served (permissive licence) · get_code("b761db2fcfa1008b")
How it ran

no invocation recorded.

SDXL Not yet run
unverified (harvested, no recorded run)
potato-kitty/EasyInv/EasyInv_SDXL/utils.py
code served (permissive licence) · get_code("a65f723dd78e582a")
How it ran

no invocation recorded.

diffusion_step Not yet run
unverified (harvested, no recorded run)
potato-kitty/EasyInv/utilsV14.py
code served (permissive licence) · get_code("f8ea771130ea35eb")
How it ran

no invocation recorded.

get_down_block Not yet run
unverified (harvested, no recorded run)
potato-kitty/EasyInv/replace_diffuser/unet_2d_blocks.py
code served (permissive licence) · get_code("e753703e70c30f2b")
How it ran

no invocation recorded.

get_up_block Not yet run
unverified (harvested, no recorded run)
potato-kitty/EasyInv/replace_diffuser/unet_2d_blocks.py
code served (permissive licence) · get_code("bcf7f25cb22d7136")
How it ran

no invocation recorded.

image2latent Not yet run
unverified (harvested, no recorded run)
potato-kitty/EasyInv/utilsV14.py
code served (permissive licence) · get_code("8a7518dbffd1c318")
How it ran

no invocation recorded.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 2 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 1 the same month, 2 later) and 8 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
retrieve_timesteps 21 other papers
8 dated earlier, 2 the same month, 11 later
2310.06313 (2023-10)
Advancing Pose-Guided Image Synthesis with Progressive Conditional Diffusion Models
2401.11708 (2024-01)
Mastering Text-to-Image Diffusion: Recaptioning, Planning, and Generating with Multimodal LLMs
2403.06976 (2024-03)
BrushNet: A Plug-and-Play Image Inpainting Model with Decomposed Dual-Branch Diffusion
and 18 more not named here
AttentionStore
brick of this paper's assembly: entry
1 other paper
1 later
2412.02237 (2024-12)
Cross-Attention Head Position Patterns Can Align with Human Visual Concepts in Text-to-Image Generative 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

This paper introduces EasyInv, an easy yet novel approach that significantly advances the field of DDIM Inversion by addressing the inherent inefficiencies and performance limitations of traditional iterative optimization methods. At the core of our EasyInv is a refined strategy for approximating inversion noise, which is pivotal for enhancing the accuracy and reliability of the inversion process. By prioritizing the initial latent state, which encapsulates rich information about the original images, EasyInv steers clear of the iterative refinement of noise items. Instead, we introduce a methodical aggregation of the latent state from the preceding time step with the current state, effectively increasing the influence of the initial latent state and mitigating the impact of noise. We illustrate that EasyInv is capable of delivering results that are either on par with or exceed those of the conventional DDIM Inversion approach, especially under conditions where the model's precision is limited or computational resources are scarce. Concurrently, our EasyInv offers an approximate threefold enhancement regarding inference efficiency over off-the-shelf iterative optimization techniques. See code at https:// github.com/potato-kitty/EasyInv.

For agents

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

get_harvested_code_for_paper("2408.05159")
get_code_for_paper("2408.05159")
have("2408.05159")

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

curl https://syntology.ai/api/ran/2408.05159.json

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

[![Syntology run record](https://syntology.ai/api/ran/2408.05159.svg)](https://syntology.ai/paper/2408.05159)

Connect an agent — have() is free.