SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2203.14486 · CVPR · 2022

Equivariant Point Cloud Analysis via Learning Orientations for Message Passing

Jian Peng, Shitong Luo, Jiaqi Guan, Jianzhu Ma, Yufeng Su, Chaoran Cheng, Jiahan Li

arXiv · PDF · Open in the Atlas

Code that ran

14 of 15 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 5 have no contract verdict; 8 failures of our instrument; 1 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/2203.14486.json · badge: /api/ran/2203.14486.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
luost26/equivariant-orientedmp canonical (classifier label)
classifier basis not recorded on the link
14 ran · 1 no recorded run
FunctionStatusWhere it lives
calculate_shape_IoU Ran
ran on a synthesized input
luost26/equivariant-orientedmp/train_partseg.py
code served (permissive licence) · get_code("73a932f7826aa350")
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 numpy as np

# Create small sample data for testing
# pred_np: predicted segmentation labels (batch_size, num_points)
# seg_np: ground truth segmentation labels (batch_size, num_points)
# num_seg_cls: number of segmentation classes

batch_size = 2
num_points = 4
num_seg_cls = 3

# Create predicted segmentation array with class labels
pred_np = np.array([
    [0, 1, 2, 0],
    [1, 2, 0, 1]
], dtype=np.int64)

# Create ground truth segmentation array with class labels
seg_np = np.array([
    [0, 1, 1, 0],
    [1, 2, 2, 1]
], dtype=np.int64)

ARGS = {
    "pred_np": pred_np,
    "seg_np": seg_np,
    "num_seg_cls": num_seg_cls
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — 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
farthest_point_sample Ran
ran; our contract draft was wrong, not the code
luost26/equivariant-orientedmp/datasets/modelnet.py
pointer only (licence: MIT); the stored copy get_code serves, from Yuzhe-Fu/FlashFPS, is pointer only (licence: NOASSERTION) · get_code("f80066a00e7156a2")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from Yuzhe-Fu/FlashFPS, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

gather Ran
ran on a synthesized input
luost26/equivariant-orientedmp/models/cls/oriented_dgcnn.py
code served (permissive licence) · get_code("8e1b540b7dbb4fd7")
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, 3, 5, 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)

# Based on the function signature and body:
# - x: (B, d, N) tensor
# - idx: (B, N, k) tensor of indices
# The function gathers features from x using idx

batch_size = 2
num_dims = 4
num_points = 3
k = 5

# Create x with shape (B, d, N)
x = torch.randn(batch_size, num_dims, num_points)

# Create idx with shape (B, N, k)
# idx should contain valid indices in range [0, num_points)
idx = torch.randint(0, num_points, (batch_size, num_points, k))

ARGS = {
    "x": x,
    "idx": idx,
}
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, 3, 5, 4) — shape (2, 3, 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
get_graph_feature Ran
ran on a synthesized input
luost26/equivariant-orientedmp/models/cls/oriented_dgcnn.py
code served (permissive licence) · get_code("d98fb5121c2ad8dd")
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, 8, 5, 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

torch.manual_seed(0)

# Construct valid arguments for get_graph_feature
batch_size = 2
num_dims = 4
num_points = 5
k = 3

# x: (B, d, N) - batch_size=2, num_dims=4, num_points=5
x = torch.randn(batch_size, num_dims, num_points)

# k: number of nearest neighbors
k_value = k

# idx: optional, (B, N, k) - if not provided, it will be computed via knn
idx = None

# x_coord: optional, coordinate tensor for knn computation
x_coord = None

ARGS = {
    "x": x,
    "k": k_value,
    "idx": idx,
    "x_coord": x_coord,
}
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, 8, 5, 3) — shape (2, 8, 5, 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
get_hierarchical_idx Ran
ran on a synthesized input
luost26/equivariant-orientedmp/modules/frame.py
code served (permissive licence) · get_code("9bad2eb9e3f5facf")
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. 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 numpy as np

# Set random seed for reproducibility
np.random.seed(0)

# Construct arguments for get_hierarchical_idx
n = 10  # total number of elements
h = [3, 5, 10]  # hierarchy levels specifying how many elements to select at each level
fps = False  # use random sampling instead of fps

ARGS = {
    "n": n,
    "h": h,
    "fps": fps
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list — a list 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
get_hierarchical_idx Ran
ran on a synthesized input
luost26/equivariant-orientedmp/models/cls/oriented_rscnn.py
code served (permissive licence) · get_code("cc0632008d5a9aef")
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. 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 numpy as np

ARGS = {
    "n": 1000,
    "h": [512, 128]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list — a list 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
knn Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
luost26/equivariant-orientedmp/models/cls/oriented_dgcnn.py
code served (permissive licence) · get_code("cdd0141594039dcb")
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 3, dtype family int, shape (2, 3, 2) 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)

# x is used with:
# - transpose(2, 1): expects 3D tensor
# - matmul with transposed version: matrix multiplication
# - sum(x ** 2, dim=1, keepdim=True): sum over dimension 1
# - topk(k=k, dim=-1): get top k indices along last dimension
# 
# Shape should be (batch, features, num_points) based on the transpose operations
# Let's use batch=2, features=4, num_points=3

x = torch.randn(2, 4, 3)

# k is the number of nearest neighbors to find
# Should be less than num_points (3), so k=2
k = 2

ARGS = {
    "x": x,
    "k": k,
}
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 3, dtype family int, shape (2, 3, 2) — shape (2, 3, 2) 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
normalize_vector Ran
ran; our contract draft was wrong, not the code
luost26/equivariant-orientedmp/modules/geometric.py
pointer only (licence: MIT); the stored copy get_code serves, from A4Bio/ADesigner, is pointer only (licence: NONE) · get_code("d26ff18f89e81ea5")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from A4Bio/ADesigner, 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.

pc_normalize Ran
ran, honoured the contract we drafted
luost26/equivariant-orientedmp/datasets/modelnet.py
pointer only (licence: MIT); the stored copy get_code serves, from ajhamdi/MVTN, is pointer only (licence: NONE) · get_code("4783fbece52f500e")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from ajhamdi/MVTN, 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.

project_v2v Ran
ran; our contract draft was wrong, not the code
luost26/equivariant-orientedmp/modules/geometric.py
code served (permissive licence) · get_code("e83f79b6a566bc97")
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
import numpy as np

torch.manual_seed(0)

# Construct valid arguments for project_v2v
v = torch.randn(2, 4, 3)  # (N=2, L=4, 3)
e = torch.randn(2, 4, 3)  # (N=2, L=4, 3)
dim = 2  # dimension to sum over

ARGS = {
    "v": v,
    "e": e,
    "dim": dim,
}
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, 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
randn Ran
ran; our contract draft was wrong, not the code
luost26/equivariant-orientedmp/modules/gvp.py
code served (permissive licence) · get_code("28e463a04d36b65d")
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

torch.manual_seed(0)

ARGS = {
    'n': 2,
    'dims': (4, 3),
    'device': 'cpu'
}
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
register_model Ran
ran; our contract draft was wrong, not the code
luost26/equivariant-orientedmp/models/cls/_registry.py
pointer only (licence: MIT); the stored copy get_code serves, from edapinenut/cbgbench, is pointer only (licence: GPL-3.0) · get_code("be3d8f2945b52b90")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from edapinenut/cbgbench, and GPL-3.0, as GitHub detects it on that repository, is not in the permissive set Syntology serves inline. 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.

safe_norm Ran
ran; our contract draft was wrong, not the code
luost26/equivariant-orientedmp/modules/geometric.py
pointer only (licence: MIT); the stored copy get_code serves, from smiles724/peptidedesign, is pointer only (licence: NONE) · get_code("e253039723a2f1f8")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from smiles724/peptidedesign, 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.

tuple_index Ran
ran; our contract draft was wrong, not the code
luost26/equivariant-orientedmp/modules/gvp.py
pointer only (licence: MIT); the stored copy get_code serves, from ai4protein/VenusX, is pointer only (licence: NOASSERTION) · get_code("38735f4bca7f8220")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from ai4protein/VenusX, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

get_model Not yet run
unverified (harvested, no recorded run)
luost26/equivariant-orientedmp/models/cls/_registry.py
code served (permissive licence) · get_code("be1d33da9ab60900")
How it ran

no invocation recorded.

Also present in other papers' code

Of 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 15 distinct functions with an identity, 10 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 3 the same month, 10 later) and 5 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
knn 52 other papers
16 dated earlier, 2 the same month, 34 later · 12 only as an assembly brick
1801.07829 (2018-01)
Dynamic Graph CNN for Learning on Point Clouds
1905.06292 (2019-05)
3D Point Cloud Generative Adversarial Network Based on Tree Structured Graph Convolutions
2005.06734 (2020-05)
Dense-Resolution Network for Point Cloud Classification and Segmentation
and 49 more not named here
pc_normalize 47 other papers
8 dated earlier, 3 the same month, 36 later
1811.07246 (2018-11)
PointConv: Deep Convolutional Networks on 3D Point Clouds
1912.12098 (2019-12)
Quaternion Equivariant Capsule Networks for 3D Point Clouds
2003.00492 (2020-03)
PointASNL: Robust Point Clouds Processing using Nonlocal Neural Networks with Adaptive Sampling
and 44 more not named here
farthest_point_sample 43 other papers
5 dated earlier, 3 the same month, 35 later
1811.07246 (2018-11)
PointConv: Deep Convolutional Networks on 3D Point Clouds
2003.00492 (2020-03)
PointASNL: Robust Point Clouds Processing using Nonlocal Neural Networks with Adaptive Sampling
2010.05272 (2020-10)
IF-Defense: 3D Adversarial Point Cloud Defense via Implicit Function based Restoration
and 40 more not named here
tuple_index 11 other papers
11 later
2309.16519 (2023-09)
AtomSurf : Surface Representation for Learning on Protein Structures
2404.11199 (2024-04)
RiboDiffusion: Tertiary Structure-based RNA Inverse Folding with Generative Diffusion Models
2408.11363 (2024-08)
ProteinGPT: Multimodal LLM for Protein Property Prediction and Structure Understanding
and 8 more not named here
randn 6 other papers
6 later
2309.16519 (2023-09)
AtomSurf : Surface Representation for Learning on Protein Structures
2409.19838 (2024-09)
Using pretrained graph neural networks with token mixers as geometric featurizers for conformational dynamics
2410.21127 (2024-10)
Retrieval-Enhanced Mutation Mastery: Augmenting Zero-Shot Prediction of Protein Language Model
and 3 more not named here
register_model 3 other papers
3 later
2204.09410 (2022-04)
Generating 3D Molecules for Target Protein Binding
2311.10908 (2023-11)
Equivariant Neural Operator Learning with Graphon Convolution
2411.18463 (2024-11)
Hotspot-Driven Peptide Design via Multi-Fragment Autoregressive Extension
normalize_vector 2 other papers
2 later
2208.06073 (2022-08)
Conditional Antibody Design as 3D Equivariant Graph Translation
2411.18463 (2024-11)
Hotspot-Driven Peptide Design via Multi-Fragment Autoregressive Extension
project_v2v 2 other papers
1 dated earlier, 1 later
2201.13299 (2022-01)
Orientation-Aware Graph Neural Networks for Protein Structure Representation Learning
2411.18463 (2024-11)
Hotspot-Driven Peptide Design via Multi-Fragment Autoregressive Extension
safe_norm 2 other papers
1 dated earlier, 1 later
2201.13299 (2022-01)
Orientation-Aware Graph Neural Networks for Protein Structure Representation Learning
2411.10618 (2024-11)
D-Flow: Multi-modality Flow Matching for D-peptide Design
get_model 1 other paper
1 later
2411.18463 (2024-11)
Hotspot-Driven Peptide Design via Multi-Fragment Autoregressive Extension

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

Equivariance has been a long-standing concern in various fields ranging from computer vision to physical modeling. Most previous methods struggle with generality, simplicity, and expressiveness -some are designed ad hoc for specific data types, some are too complex to be accessible, and some sacrifice flexible transformations. In this work, we propose a novel and simple framework to achieve equivariance for point cloud analysis based on the message passing (graph neural network) scheme. We find the equivariant property could be obtained by introducing an orientation for each point to decouple the relative position for each point from the global pose of the entire point cloud. Therefore, we extend current message passing networks with a module that learns orientations for each point. Before aggregating information from the neighbors of a point, the networks transforms the neighbors' coordinates based on the point's learned orientations. We provide formal proofs to show the equivariance of the proposed framework. Empirically, we demonstrate that our proposed method is competitive on both point cloud analysis and physical modeling tasks. Code is available at https://github.com/ luost26/Equivariant-OrientedMP.

For agents

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

get_harvested_code_for_paper("2203.14486")
get_code_for_paper("2203.14486")
have("2203.14486")

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

curl https://syntology.ai/api/ran/2203.14486.json

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

[![Syntology run record](https://syntology.ai/api/ran/2203.14486.svg)](https://syntology.ai/paper/2203.14486)

Connect an agent — have() is free.