SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2501.18616 · ICLR · 2025

STAMP: Scalable Task And Model-agnostic Collaborative Perception

Jiachen Li, Zhiwen Fan, Xiangbo Gao, Runsheng Xu, Ziran Wang, Zhengzhong Tu

arXiv · PDF · Open in the Atlas

Code that ran

10 of 11 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; 10 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/2501.18616.json · badge: /api/ran/2501.18616.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
taco-group/STAMP canonical (classifier label)
classifier basis not recorded on the link
10 ran · 1 no recorded run
FunctionStatusWhere it lives
add_sin_difference Ran
ran; our contract draft was wrong, not the code
taco-group/STAMP/opencood/loss/ciassd_loss.py
code served (permissive licence) · get_code("725b188cee8a1577")
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 4, 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

torch.manual_seed(0)

# Create two small tensors representing boxes
# Each box has coordinates and an angle (last dimension)
# Shape: (batch_size, num_boxes, features)
# Features: [x, y, z, angle] - so 4 features total
boxes1 = torch.randn(2, 3, 4)
boxes2 = torch.randn(2, 3, 4)

ARGS = {
    "boxes1": boxes1,
    "boxes2": boxes2,
}
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 4, 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
clip_sigmoid Ran
ran; our contract draft was wrong, not the code
taco-group/STAMP/opencood/loss/center_point_loss.py
pointer only (licence: MIT); the stored copy get_code serves, from yifanlu0227/HEAL, is pointer only (licence: NOASSERTION) · get_code("9163c6bd787ad901")
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 yifanlu0227/HEAL, 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.

gaussian_2d Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
taco-group/STAMP/opencood/loss/center_point_loss.py
pointer only (licence: MIT); the stored copy get_code serves, from yifanlu0227/HEAL, is pointer only (licence: NOASSERTION) · get_code("ea036153b550642a")
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 yifanlu0227/HEAL, 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.

gaussian_focal_loss Ran
ran; our contract draft was wrong, not the code
taco-group/STAMP/opencood/loss/center_point_loss.py
pointer only (licence: MIT); the stored copy get_code serves, from yifanlu0227/HEAL, is pointer only (licence: NOASSERTION) · get_code("17eaeb2d97f99989")
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 yifanlu0227/HEAL, 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_direction_target Ran
ran; our contract draft was wrong, not the code
taco-group/STAMP/opencood/loss/ciassd_loss.py
code served (permissive licence) · get_code("71d71587ce799f91")
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, 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)

# reg_targets shape: (B, H*W*2, 7)
# B=2 (batch_size), H*W*2=4 (num_anchors), 7 (last dimension for rotation + others)
reg_targets = torch.randn(2, 4, 7)

# anchors shape: (H*W*2, 7) or can be reshaped to (H, W, 2, 7)
# We'll use (4, 7) to match the middle dimension of reg_targets
anchors = torch.randn(4, 7)

ARGS = {
    "reg_targets": reg_targets,
    "anchors": anchors,
    "one_hot": True,
    "dir_offset": 0.0,
}
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 Tensor, ndim 3, dtype family float, shape (2, 4, 2) — shape (2, 4, 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
indices_to_dense_vector Ran
ran; our contract draft was wrong, not the code
taco-group/STAMP/opencood/loss/fpvrcnn_loss.py
pointer only (licence: MIT); the stored copy get_code serves, from CrazyShout/INSTINCT, is pointer only (licence: NOASSERTION) · get_code("3acf861a2d1c5a22")
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 CrazyShout/INSTINCT, 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.

one_hot_f Ran
ran; our contract draft was wrong, not the code
taco-group/STAMP/opencood/loss/ciassd_loss.py
code served (permissive licence) · get_code("cbb724502b77baff")
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, 5) 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)

# Create a small tensor with class indices
# Shape: (2, 4) - batch size 2, 4 elements to encode
tensor = torch.tensor([[0, 1, 2], [1, 0, 3]], dtype=torch.long)

# Depth of one-hot encoding (number of classes)
depth = 5

# Dimension along which to scatter (default -1 means last dimension)
dim = -1

# Value to use for "on" positions in one-hot encoding
on_value = 1.0

# Data type for output tensor
dtype = torch.float32

ARGS = {
    "tensor": tensor,
    "depth": depth,
    "dim": dim,
    "on_value": on_value,
    "dtype": dtype,
}
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, 3, 5) — shape (2, 3, 5) 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
one_hot_f Ran
ran; our contract draft was wrong, not the code
taco-group/STAMP/opencood/loss/point_pillar_loss.py
pointer only (licence: MIT); the stored copy get_code serves, from CrazyShout/INSTINCT, is pointer only (licence: NOASSERTION) · get_code("3508d3e1819272d8")
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 CrazyShout/INSTINCT, 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.

softmax_cross_entropy_with_logits Ran
ran; our contract draft was wrong, not the code
taco-group/STAMP/opencood/loss/point_pillar_loss.py
pointer only (licence: MIT); the stored copy get_code serves, from CrazyShout/INSTINCT, is pointer only (licence: NOASSERTION) · get_code("44590278c8d435af")
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 CrazyShout/INSTINCT, 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.

weighted_smooth_l1_loss Ran
ran; our contract draft was wrong, not the code
taco-group/STAMP/opencood/loss/point_pillar_loss.py
pointer only (licence: MIT); the stored copy get_code serves, from CrazyShout/INSTINCT, is pointer only (licence: NOASSERTION) · get_code("3318bc2d0e244104")
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 CrazyShout/INSTINCT, 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.

weighted_sigmoid_binary_cross_entropy Not yet run
unverified (harvested, no recorded run)
taco-group/STAMP/opencood/loss/fpvrcnn_loss.py
pointer only (licence: MIT); the stored copy get_code serves, from CrazyShout/INSTINCT, is pointer only (licence: NOASSERTION) · get_code("cbe2c496241c0635")
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 CrazyShout/INSTINCT, 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. No invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 11 are also present in other papers whose code we hold (11 with at least one paper dated earlier, 0 the same month, 8 later) and 0 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
indices_to_dense_vector 10 other papers
8 dated earlier, 2 later
1903.11027 (2019-03)
nuScenes: A multimodal dataset for autonomous driving
2207.08560 (2022-07)
Latency-Aware Collaborative Perception
2303.09495 (2023-03)
Among Us: Adversarially Robust Collaborative Perception by Consensus
and 7 more not named here
add_sin_difference 7 other papers
5 dated earlier, 2 later
2307.11514 (2023-07)
CORE: Cooperative Reconstruction for Multi-Agent Perception
2310.08117 (2023-10)
DUSA: Decoupled Unsupervised Sim2Real Adaptation for Vehicle-to-Everything Collaborative Perception
2401.13964 (2024-01)
An Extensible Framework for Open Heterogeneous Collaborative Perception
and 4 more not named here
get_direction_target 7 other papers
5 dated earlier, 2 later
2307.11514 (2023-07)
CORE: Cooperative Reconstruction for Multi-Agent Perception
2310.08117 (2023-10)
DUSA: Decoupled Unsupervised Sim2Real Adaptation for Vehicle-to-Everything Collaborative Perception
2401.13964 (2024-01)
An Extensible Framework for Open Heterogeneous Collaborative Perception
and 4 more not named here
one_hot_f 7 other papers
5 dated earlier, 2 later
2307.11514 (2023-07)
CORE: Cooperative Reconstruction for Multi-Agent Perception
2310.08117 (2023-10)
DUSA: Decoupled Unsupervised Sim2Real Adaptation for Vehicle-to-Everything Collaborative Perception
2401.13964 (2024-01)
An Extensible Framework for Open Heterogeneous Collaborative Perception
and 4 more not named here
weighted_sigmoid_binary_cross_entropy 7 other papers
5 dated earlier, 2 later
2307.11514 (2023-07)
CORE: Cooperative Reconstruction for Multi-Agent Perception
2310.08117 (2023-10)
DUSA: Decoupled Unsupervised Sim2Real Adaptation for Vehicle-to-Everything Collaborative Perception
2401.13964 (2024-01)
An Extensible Framework for Open Heterogeneous Collaborative Perception
and 4 more not named here
gaussian_2d 6 other papers
6 dated earlier
2006.11275 (2020-06)
Center-based 3D Object Detection and Tracking
2203.10642 (2022-03)
FUTR3D: A Unified Sensor Fusion Framework for 3D Detection
2205.13542 (2022-05)
BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representation
and 3 more not named here
one_hot_f 4 other papers
3 dated earlier, 1 later
2310.14702 (2023-10)
BM2CP: Efficient Collaborative Perception with LiDAR-Camera Modalities
2401.13964 (2024-01)
An Extensible Framework for Open Heterogeneous Collaborative Perception
2411.08402 (2024-11)
V2X-R: Cooperative LiDAR-4D Radar Fusion with Denoising Diffusion for 3D Object Detection
and 1 more not named here
softmax_cross_entropy_with_logits 4 other papers
3 dated earlier, 1 later
2310.14702 (2023-10)
BM2CP: Efficient Collaborative Perception with LiDAR-Camera Modalities
2401.13964 (2024-01)
An Extensible Framework for Open Heterogeneous Collaborative Perception
2411.08402 (2024-11)
V2X-R: Cooperative LiDAR-4D Radar Fusion with Denoising Diffusion for 3D Object Detection
and 1 more not named here
weighted_smooth_l1_loss 4 other papers
3 dated earlier, 1 later
2310.14702 (2023-10)
BM2CP: Efficient Collaborative Perception with LiDAR-Camera Modalities
2401.13964 (2024-01)
An Extensible Framework for Open Heterogeneous Collaborative Perception
2411.08402 (2024-11)
V2X-R: Cooperative LiDAR-4D Radar Fusion with Denoising Diffusion for 3D Object Detection
and 1 more not named here
clip_sigmoid 1 other paper
1 dated earlier
2401.13964 (2024-01)
An Extensible Framework for Open Heterogeneous Collaborative Perception
gaussian_focal_loss 1 other paper
1 dated earlier
2401.13964 (2024-01)
An Extensible Framework for Open Heterogeneous Collaborative Perception

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

Perception is a crucial component of autonomous driving systems. However, single-agent setups often face limitations due to sensor constraints, especially under challenging conditions like severe occlusion, adverse weather, and long-range object detection. Multi-agent collaborative perception (CP) offers a promising solution that enables communication and information sharing between connected vehicles. Yet, the heterogeneity among agents-in terms of sensors, models, and tasks-significantly hinders effective and efficient cross-agent collaboration. To address these challenges, we propose STAMP, a scalable task-and model-agnostic collaborative perception framework tailored for heterogeneous agents. STAMP utilizes lightweight adapter-reverter pairs to transform Bird's Eye View (BEV) features between agent-specific domains and a shared protocol domain, facilitating efficient feature sharing and fusion while minimizing computational overhead. Moreover, our approach enhances scalability, preserves model security, and accommodates a diverse range of agents. Extensive experiments on both simulated (OPV2V) and real-world (V2V4Real) datasets demonstrate that STAMP achieves comparable or superior accuracy to state-of-the-art models with significantly reduced computational costs. As the first-of-its-kind task-and model-agnostic collaborative perception framework, STAMP aims to advance research in scalable and secure mobility systems, bringing us closer to Level 5 autonomy. Our project page is at https://xiangbogaobarry.github.io/STAMP and the code is available at https://github.com/taco-group/STAMP.

For agents

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

get_harvested_code_for_paper("2501.18616")
get_code_for_paper("2501.18616")
have("2501.18616")

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

curl https://syntology.ai/api/ran/2501.18616.json

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

[![Syntology run record](https://syntology.ai/api/ran/2501.18616.svg)](https://syntology.ai/paper/2501.18616)

Connect an agent — have() is free.