SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2401.02292 · AAAI · 2024

GridFormer: Point-Grid Transformer for Surface Reconstruction

Yu-Shen Liu, Ge Gao, Yudong Liu, Ming Gu, Shengtao Li

arXiv · PDF · Open in the Atlas

Code that ran

9 of 9 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (6 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 1 violated it and 6 have no contract verdict; 2 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/2401.02292.json · badge: /api/ran/2401.02292.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
list17/gridformer canonical (classifier label)
classifier basis: census
2 ran · 0 no recorded run
list17/GridFormer community
Papers with Code archive
7 ran · 0 no recorded run
FunctionStatusWhere it lives
AddNorm Ran
ran on a synthesized input · constructed an object; no result was computed
list17/GridFormer/src/transformer.py
code served (permissive licence) · get_code("15bb20dc0259acf6")
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, AddNorm: 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)

# The AddNorm class expects:
# 1. norm_shape: shape parameter for LayerNorm
# 2. dropout: dropout probability

# Looking at the forward method:
# - Takes X and Y as inputs
# - Applies dropout to Y
# - Adds X and dropout(Y)
# - Applies layer norm to the result
# This means X and Y should have compatible shapes for addition

# Create small tensors for testing
batch_size = 2
feature_dim = 4

# X and Y should be compatible for addition
# They'll be added together, so same shape
X = torch.randn(batch_size, feature_dim)
Y = torch.randn(batch_size, feature_dim)

# norm_shape should match the last dimension(s) that LayerNorm will normalize
# Since we're adding tensors and then normalizing, norm_shape should be feature_dim
norm_shape = feature_dim

# dropout probability (value between 0 and 1)
dropout_prob = 0.1

ARGS = {
    "norm_shape": norm_shape,
    "dropout": dropout_prob,
}
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 AddNorm — 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 AddNorm 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
DecoderBlock Ran
ran on a synthesized input · constructed an object; no result was computed
list17/GridFormer/src/transformer.py
code served (permissive licence) · get_code("43654c002f23f271")
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, DecoderBlock: 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
from torch import nn

torch.manual_seed(0)

# Define ARGS dict with all parameters for DecoderBlock
ARGS = {
    'q_size': 8,
    'k_size': 8,
    'v_size': 8,
    'num_hiddens': 16,
    'ffn_num_input': 16,
    'ffn_num_hiddens': 32,
    'num_heads': 2,
    'dropout': 0.1,
}
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 DecoderBlock — 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 DecoderBlock 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
DotProductAttention Ran
ran on a synthesized input · constructed an object; no result was computed
list17/GridFormer/src/transformer.py
code served (permissive licence) · get_code("09f1ff3d2cceb51e")
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, DotProductAttention: 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
from torch import nn
import math

torch.manual_seed(0)

# Create small tensors for the attention mechanism
batch_size = 2
seq_len = 3
d_model = 4

# queries: shape (batch_size, seq_len, d_model)
queries = torch.randn(batch_size, seq_len, d_model)

# keys: shape (batch_size, seq_len, d_model)
keys = torch.randn(batch_size, seq_len, d_model)

# values: shape (batch_size, seq_len, d_model)
values = torch.randn(batch_size, seq_len, d_model)

# valid_lens: shape (batch_size,) - specify valid sequence lengths for masking
valid_lens = torch.tensor([2, 3])

# Create DotProductAttention instance
dropout_rate = 0.1

ARGS = {
    'dropout': dropout_rate,
    'num_heads': 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 DotProductAttention — 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 DotProductAttention 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
MultiHeadAttention Ran
ran on a synthesized input · constructed an object; no result was computed
list17/GridFormer/src/transformer.py
code served (permissive licence) · get_code("45b7b846b99a837c")
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, MultiHeadAttention: 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
from torch import nn
import math

torch.manual_seed(0)

# Parameters for MultiHeadAttention
q_size = 4
k_size = 4
v_size = 4
num_hiddens = 8
num_heads = 2
dropout = 0.1
bias = False

ARGS = {
    "q_size": q_size,
    "k_size": k_size,
    "v_size": v_size,
    "num_hiddens": num_hiddens,
    "num_heads": num_heads,
    "dropout": dropout,
    "bias": bias,
}
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 MultiHeadAttention — 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 MultiHeadAttention 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
PositionWiseFFN Ran
ran on a synthesized input · constructed an object; no result was computed
list17/GridFormer/src/transformer.py
code served (permissive licence) · get_code("e5915e3055401b5f")
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, PositionWiseFFN: 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
from torch import nn

torch.manual_seed(0)

# The PositionWiseFFN class takes three parameters:
# - ffn_num_input: input dimension
# - ffn_num_hiddens: hidden layer dimension
# - ffn_num_outputs: output dimension

# These should be distinct integers to allow shape probing
ffn_num_input = 4
ffn_num_hiddens = 8
ffn_num_outputs = 3

ARGS = {
    "ffn_num_input": ffn_num_input,
    "ffn_num_hiddens": ffn_num_hiddens,
    "ffn_num_outputs": ffn_num_outputs,
}
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 PositionWiseFFN — 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 PositionWiseFFN 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
TransformerDecoder Ran
ran on a synthesized input · constructed an object; no result was computed
list17/GridFormer/src/transformer.py
code served (permissive licence) · get_code("8a7c1ecdc48ac3e3")
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, TransformerDecoder: 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
from torch import nn
import math

torch.manual_seed(0)

ARGS = {
    'q_size': 4,
    'k_size': 5,
    'v_size': 6,
    'output_size': 7,
    'num_hiddens': 8,
    'ffn_num_input': 8,
    'ffn_num_hiddens': 16,
    'num_heads': 2,
    'num_blks': 1,
    'dropout': 0.1,
}
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 TransformerDecoder — 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 TransformerDecoder 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_box_corners_3d Ran
ran; our contract draft was wrong, not the code
list17/gridformer/modules/frustum.py
code served (permissive licence) · get_code("3ae68d06c43d9791")
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 3, dtype family float, shape (2, 3, 8) 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)

# Create valid arguments for get_box_corners_3d
N = 2  # batch size

# centers: FloatTensor[N, 3] - coordinates of box centers
centers = torch.randn(N, 3, dtype=torch.float32)

# headings: FloatTensor[N, ] - heading angles in radians
headings = torch.randn(N, dtype=torch.float32)

# sizes: FloatTensor[N, 3] - box sizes (length, width, height)
sizes = torch.randn(N, 3, dtype=torch.float32) + 2.0  # avoid near-zero sizes

# with_flip: bool - whether to return flipped box
with_flip = False

ARGS = {
    "centers": centers,
    "headings": headings,
    "sizes": sizes,
    "with_flip": with_flip,
}
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 3, dtype family float, shape (2, 3, 8) — shape (2, 3, 8) 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
is_url Ran
ran, violated the contract we drafted
list17/gridformer/src/checkpoints.py
code served (permissive licence) · get_code("e39d881796e8d67e")
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 bool, ndim 0, dtype family float, 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 urllib

def is_url(url):
    scheme = urllib.parse.urlparse(url).scheme
    return scheme in ('http', 'https')

ARGS = {
    'url': 'https://example.com'
}
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 bool, ndim 0, dtype family float, 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
masked_softmax Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
list17/GridFormer/src/transformer.py
code served (permissive licence) · get_code("240db2e22154b1de")
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 3, dtype family float, shape (2, 4, 3) 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
from torch import nn

torch.manual_seed(0)

# Create a batch of attention scores/logits
# Shape: (batch_size=2, seq_len=4, hidden_dim=3)
X = torch.randn(2, 4, 3)

# Create valid_lens indicating actual sequence lengths
# For batch of 2, with sequence lengths 3 and 4
valid_lens = torch.tensor([3, 4])

ARGS = {
    "X": X,
    "valid_lens": valid_lens
}
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 3, dtype family float, shape (2, 4, 3) — shape (2, 4, 3) 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 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 1 the same month, 0 later) and 7 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
is_url 12 other papers
11 dated earlier, 1 the same month
1912.07372 (2019-12)
Differentiable Volumetric Rendering: Learning Implicit 3D Representations without 3D Supervision
2001.05119 (2020-01)
Learning multiview 3D point cloud registration
2003.04618 (2020-03)
Convolutional Occupancy Networks
and 9 more not named here
get_box_corners_3d 2 other papers
2 dated earlier
1907.03739 (2019-07)
Point-Voxel CNN for Efficient 3D Deep Learning
2006.03677 (2020-06)
Visual Transformers: Token-based Image Representation and Processing for Computer Vision

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

Implicit neural networks have emerged as a crucial technology in 3D surface reconstruction. To reconstruct continuous surfaces from discrete point clouds, encoding the input points into regular grid features (plane or volume) has been commonly employed in existing approaches. However, these methods typically use the grid as an index for uniformly scattering point features. Compared with the irregular point features, the regular grid features may sacrifice some reconstruction details but improve efficiency. To take full advantage of these two types of features, we introduce a novel and high-efficiency attention mechanism between the grid and point features named Point-Grid Transformer (GridFormer). This mechanism treats the grid as a transfer point connecting the space and point cloud. Our method maximizes the spatial expressiveness of grid features and maintains computational efficiency. Furthermore, optimizing predictions over the entire space could potentially result in blurred boundaries. To address this issue, we further propose a boundary optimization strategy incorporating margin binary cross-entropy loss and boundary sampling. This approach enables us to achieve a more precise representation of the object structure. Our experiments validate that our method is effective and outperforms the state-of-the-art approaches under widely used benchmarks by producing more precise geometry reconstructions. The code is available at https://github.com/list17/GridFormer.

For agents

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

get_harvested_code_for_paper("2401.02292")
get_code_for_paper("2401.02292")
have("2401.02292")

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

curl https://syntology.ai/api/ran/2401.02292.json

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

[![Syntology run record](https://syntology.ai/api/ran/2401.02292.svg)](https://syntology.ai/paper/2401.02292)

Connect an agent — have() is free.