SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2111.00643 · NeurIPS · 2021

Learning Distilled Collaboration Graph for Multi-Agent Perception

Yiming Li, Siheng Chen, Pengxiang Wu, Shunli Ren, Chen Feng, Wenjun Zhang

arXiv · PDF · Open in the Atlas

Code that ran

4 of 6 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (4 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 4 have no contract verdict; 0 failures of our instrument; 2 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/2111.00643.json · badge: /api/ran/2111.00643.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
coperception/coperception community
Papers with Code archive
4 ran · 2 no recorded run
FunctionStatusWhere it lives
ClassificationHead Ran
ran on a synthesized input · constructed an object; no result was computed
coperception/coperception/coperception/models/det/DiscoNet.py
code served (permissive licence) · get_code("2eefc49e3fbd30e1")
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, ClassificationHead: 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)

class Config:
    def __init__(self):
        self.category_num = 5
        self.use_map = True
        self.use_vis = True
        self.anchor_size = [(1, 1), (2, 2), (3, 3)]

config = Config()

ARGS = {
    'config': config
}
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 ClassificationHead — 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 ClassificationHead 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
DetModelBase Ran
ran on a synthesized input · constructed an object; no result was computed
coperception/coperception/coperception/models/det/DiscoNet.py
code served (permissive licence) · get_code("642d741e6a6dde66")
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, DetModelBase: 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
import numpy as np

torch.manual_seed(0)

class Config:
    def __init__(self):
        self.category_num = 3
        self.use_map = True
        self.use_vis = False
        self.anchor_size = [(1.0, 2.0), (1.5, 2.5)]
        self.box_code_size = 4
        self.pred_len = 5
        self.only_det = False
        self.binary = True
        self.motion_state = True

config = Config()

ARGS = {
    'config': config,
    'layer': 2,
    'in_channels': 13,
    'kd_flag': True,
    'p_com_outage': 0.1,
    'num_agent': 4,
    'only_v2i': False
}
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 DetModelBase — 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 DetModelBase 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
PixelWeightedFusionSoftmax Ran
ran on a synthesized input · constructed an object; no result was computed
coperception/coperception/coperception/models/det/DiscoNet.py
code served (permissive licence) · get_code("bf62f79dbe83527d")
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, PixelWeightedFusionSoftmax: 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)

class PixelWeightedFusionSoftmax(nn.Module):

    def __init__(self, channel):
        super(PixelWeightedFusionSoftmax, self).__init__()
        self.conv1_1 = nn.Conv2d(channel * 2, 128, kernel_size=1, stride=1, padding=0)
        self.bn1_1 = nn.BatchNorm2d(128)
        self.conv1_2 = nn.Conv2d(128, 32, kernel_size=1, stride=1, padding=0)
        self.bn1_2 = nn.BatchNorm2d(32)
        self.conv1_3 = nn.Conv2d(32, 8, kernel_size=1, stride=1, padding=0)
        self.bn1_3 = nn.BatchNorm2d(8)
        self.conv1_4 = nn.Conv2d(8, 1, kernel_size=1, stride=1, padding=0)

    def forward(self, x):
        x = x.view(-1, x.size(-3), x.size(-2), x.size(-1))
        import torch.nn.functional as F
        x_1 = F.relu(self.bn1_1(self.conv1_1(x)))
        x_1 = F.relu(self.bn1_2(self.conv1_2(x_1)))
        x_1 = F.relu(self.bn1_3(self.conv1_3(x_1)))
        x_1 = F.relu(self.conv1_4(x_1))
        return x_1

channel = 4
model = PixelWeightedFusionSoftmax(channel=channel)
batch_size = 2
height = 3
width = 5
x = torch.randn(batch_size, channel * 2, height, width)

ARGS = {
    'channel': channel
}
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 PixelWeightedFusionSoftmax — 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 PixelWeightedFusionSoftmax 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
SingleRegressionHead Ran
ran on a synthesized input · constructed an object; no result was computed
coperception/coperception/coperception/models/det/DiscoNet.py
code served (permissive licence) · get_code("9d329a535c4071da")
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, SingleRegressionHead: 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)

class Config:
    def __init__(self):
        self.use_map = False
        self.use_vis = False
        self.anchor_size = [1, 2, 3]
        self.box_code_size = 4
        self.only_det = True
        self.pred_len = 5
        self.binary = True

config = Config()

ARGS = {
    "config": config
}
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 SingleRegressionHead — 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 SingleRegressionHead 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
DiscoNet Not yet run
unverified (harvested, no recorded run)
coperception/coperception/coperception/models/det/DiscoNet.py
code served (permissive licence) · get_code("20f1d033168b49c9")
How it ran

no invocation recorded.

IntermediateModelBase Not yet run
unverified (harvested, no recorded run)
coperception/coperception/coperception/models/det/DiscoNet.py
code served (permissive licence) · get_code("b69c9899c17318e1")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 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
ClassificationHead
brick of this paper's assembly: helper
1 other paper
1 later
2303.12400 (2023-03)
UMC: A Unified Bandwidth-efficient and Multi-resolution based Collaborative Perception Framework

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

To promote better performance-bandwidth trade-off for multi-agent perception, we propose a novel distilled collaboration graph (DiscoGraph) to model trainable, pose-aware, and adaptive collaboration among agents. Our key novelties lie in two aspects. First, we propose a teacher-student framework to train DiscoGraph via knowledge distillation. The teacher model employs an early collaboration with holistic-view inputs; the student model is based on intermediate collaboration with single-view inputs. Our framework trains DiscoGraph by constraining postcollaboration feature maps in the student model to match the correspondences in the teacher model. Second, we propose a matrix-valued edge weight in DiscoGraph. In such a matrix, each element reflects the inter-agent attention at a specific spatial region, allowing an agent to adaptively highlight the informative regions. During inference, we only need to use the student model named as the distilled collaboration network (DiscoNet). Attributed to the teacher-student framework, multiple agents with the shared DiscoNet could collaboratively approach the performance of a hypothetical teacher model with a holistic view. Our approach is validated on V2X-Sim 1.0, a large-scale multi-agent perception dataset that we synthesized using CARLA and SUMO co-simulation. Our quantitative and qualitative experiments in multi-agent 3D object detection show that DiscoNet could not only achieve a better performance-bandwidth trade-off than the state-of-the-art collaborative perception methods, but also bring more straightforward design rationale. Our code is available on https://github.com/ai4ce/DiscoNet.

For agents

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

get_harvested_code_for_paper("2111.00643")
get_code_for_paper("2111.00643")
have("2111.00643")

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

curl https://syntology.ai/api/ran/2111.00643.json

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

[![Syntology run record](https://syntology.ai/api/ran/2111.00643.svg)](https://syntology.ai/paper/2111.00643)

Connect an agent — have() is free.