SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1805.06771 · 2018

Convolutional Social Pooling for Vehicle Trajectory Prediction

arXiv · PDF · Open in the Atlas

Code that ran

6 of 9 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 5 have no contract verdict; 1 failure of our instrument; 3 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/1805.06771.json · badge: /api/ran/1805.06771.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
nachiket92/conv-social-pooling canonical (classifier label)
classifier basis not recorded on the link
2 ran · 1 no recorded run
m-hasan-n/pooling community (classifier label)
classifier basis not recorded on the link
4 ran · 2 no recorded run
FunctionStatusWhere it lives
compute_nll_mat_red Ran
ran on a synthesized input
m-hasan-n/pooling/utils.py
code served (permissive licence) · get_code("65b213ada32f2714")
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 2, dtype family float, shape (2, 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)

# y_pred shape: (batch_size, seq_len, 7) where 7 = [muX, muY, muTh, sigX, sigY, sigTh, rho]
batch_size = 2
seq_len = 4
y_pred = torch.randn(batch_size, seq_len, 7)

# y_gt shape: (batch_size, seq_len, 3) where 3 = [x, y, th]
y_gt = torch.randn(batch_size, seq_len, 3)

ARGS = {
    "y_pred": y_pred,
    "y_gt": y_gt
}
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 2, dtype family float, shape (2, 4) — shape (2, 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
cs_pooling Ran
ran on a synthesized input
m-hasan-n/pooling/pooling_module.py
code served (permissive licence) · get_code("7ecd03f90a465fb6")
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 2, dtype family float, shape (16, 16) 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 torch.nn as nn

torch.manual_seed(0)

class SocNet(nn.Module):
    def __init__(self, soc_embedding_size=16):
        super().__init__()
        self.soc_embedding_size = soc_embedding_size
        # soc_conv: takes input and outputs intermediate features
        self.soc_conv = nn.Conv1d(4, 8, kernel_size=1)
        # conv_3x1: takes 8 channels, outputs 8 channels
        self.conv_3x1 = nn.Conv1d(8, 8, kernel_size=3, padding=1)
        # soc_maxpool: pools to desired embedding size
        self.soc_maxpool = nn.AdaptiveMaxPool1d(soc_embedding_size)
        self.leaky_relu = nn.LeakyReLU()

net = SocNet(soc_embedding_size=16)

# Create a small input tensor: batch_size=2, channels=4, length=5
# The function applies convolutions and pooling to this
soc_enc = torch.randn(2, 4, 5)

ARGS = {
    "net": net,
    "soc_enc": soc_enc,
}
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 2, dtype family float, shape (16, 16) — shape (16, 16) 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
maskedNLL Ran
ran on a synthesized input
nachiket92/conv-social-pooling/utils.py
code served (permissive licence) · get_code("c411178a472256c2")
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 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 torch

torch.manual_seed(0)

# y_pred should have shape (batch, seq_len, 5)
# It contains [muX, muY, sigX, sigY, rho] at dimension 2
y_pred = torch.randn(2, 4, 5)

# y_gt should have shape (batch, seq_len, 2)
# It contains [x, y] at dimension 2
y_gt = torch.randn(2, 4, 2)

# mask should have shape (batch, seq_len, 2)
# It's used to weight the loss and for normalization
mask = torch.ones(2, 4, 2)

ARGS = {
    'y_pred': y_pred,
    'y_gt': y_gt,
    'mask': mask
}
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 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
maskedNLL Ran
ran on a synthesized input
m-hasan-n/pooling/utils.py
code served (permissive licence) · get_code("b99f681b0c923e09")
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 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 torch

torch.manual_seed(0)

# For input_dim == 5 case
batch_size = 2
seq_len = 4

# y_pred with input_dim=5: (batch, seq, 5)
y_pred_5 = torch.randn(batch_size, seq_len, 5)
# Ensure rho is in valid range (-1, 1) for the formula
y_pred_5[:, :, 4] = torch.sigmoid(y_pred_5[:, :, 4]) * 0.9 - 0.45

# y_gt with 2D coordinates: (batch, seq, 2)
y_gt_5 = torch.randn(batch_size, seq_len, 2)

# mask: (batch, seq, 2)
mask_5 = torch.ones(batch_size, seq_len, 2)

# For input_dim == 7 case
# y_pred with input_dim=7: (batch, seq, 7)
y_pred_7 = torch.randn(batch_size, seq_len, 7)
# Ensure sigX, sigY, sigTh are positive
y_pred_7[:, :, 3] = torch.abs(y_pred_7[:, :, 3]) + 0.1  # sigX
y_pred_7[:, :, 4] = torch.abs(y_pred_7[:, :, 4]) + 0.1  # sigY
y_pred_7[:, :, 5] = torch.abs(y_pred_7[:, :, 5]) + 0.1  # sigTh
# Ensure rho is in valid range
y_pred_7[:, :, 6] = torch.sigmoid(y_pred_7[:, :, 6]) * 0.9 - 0.45

# y_gt with 3D coordinates: (batch, seq, 3)
y_gt_7 = torch.randn(batch_size, seq_len, 3)

# mask: (batch, seq, 3)
mask_7 = torch.ones(batch_size, seq_len, 3)

# Test with input_dim == 5
ARGS = {
    'y_pred': y_pred_5,
    'y_gt': y_gt_5,
    'mask': mask_5
}
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 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
outputActivation Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
nachiket92/conv-social-pooling/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from duanwenchang/MAVEN-T, is pointer only (licence: NONE) · get_code("ec790045da84df86")
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 duanwenchang/MAVEN-T, 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.

outputActivation Ran
ran on a synthesized input
m-hasan-n/pooling/utils.py
code served (permissive licence) · get_code("a3e203d5da758f6d")
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)

# Test case 1: x with shape[2] == 5
x_5 = torch.randn(2, 3, 5)

# Test case 2: x with shape[2] == 7
x_7 = torch.randn(2, 3, 7)

# Use the first test case as the main argument
ARGS = {
    "x": x_5
}
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
maskedNLLTest Not yet run
unverified (harvested, no recorded run)
nachiket92/conv-social-pooling/utils.py
code served (permissive licence) · get_code("e928feb47e359b29")
How it ran

no invocation recorded.

nbrs_pooling Not yet run
unverified (harvested, no recorded run)
m-hasan-n/pooling/pooling_module.py
code served (permissive licence) · get_code("4204ae7470ca49a9")
How it ran

no invocation recorded.

s_pooling Not yet run
unverified (harvested, no recorded run)
m-hasan-n/pooling/pooling_module.py
code served (permissive licence) · get_code("bd536c1b61d6b4eb")
How it ran

no invocation recorded.

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, 3 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 3 later) and 6 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
outputActivation 2 other papers
2 later
2107.03663 (2021-07)
Graph and Recurrent Neural Network-based Vehicle Trajectory Prediction For Highway Driving
2604.10169 (2026-04)
MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
maskedNLL 1 other paper
1 later
2107.03663 (2021-07)
Graph and Recurrent Neural Network-based Vehicle Trajectory Prediction For Highway Driving
maskedNLLTest 1 other paper
1 later
2107.03663 (2021-07)
Graph and Recurrent Neural Network-based Vehicle Trajectory Prediction For Highway Driving

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

Forecasting the motion of surrounding vehicles is a critical ability for an autonomous vehicle deployed in complex traffic. Motion of all vehicles in a scene is governed by the traffic context, i.e., the motion and relative spatial configuration of neighboring vehicles. In this paper we propose an LSTM encoder-decoder model that uses convolutional social pooling as an improvement to social pooling layers for robustly learning interdependencies in vehicle motion. Additionally, our model outputs a multi-modal predictive distribution over future trajectories based on maneuver classes. We evaluate our model using the publicly available NGSIM US-101 and I-80 datasets. Our results show improvement over the state of the art in terms of RMS values of prediction error and negative log-likelihoods of true future trajectories under the model's predictive distribution. We also present a qualitative analysis of the model's predicted distributions for various traffic scenarios.

For agents

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

get_harvested_code_for_paper("1805.06771")
get_code_for_paper("1805.06771")
have("1805.06771")

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

curl https://syntology.ai/api/ran/1805.06771.json

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

[![Syntology run record](https://syntology.ai/api/ran/1805.06771.svg)](https://syntology.ai/paper/1805.06771)

Connect an agent — have() is free.